An off-line two-dimensional analytical procedure for determination of polcyclic aromatic hydrocarbons in smoke aerosol

(1)

An off-line two-dimensional analytical procedure for

determination of polcyclic aromatic hydrocarbons in smoke

aerosol

Citation for published version (APA):

Claessens, H. A., & Lammerts van Bueren, L. G. D. (1987). An off-line two-dimensional analytical procedure for determination of polcyclic aromatic hydrocarbons in smoke aerosol. HRC & CC, Journal of High Resolution Chromatography and Chromatography Communications, 10(6), 342-347.

https://doi.org/10.1002/jhrc.1240100605

DOI:

10.1002/jhrc.1240100605

Document status and date: Published: 01/01/1987

Document Version:

Publisher’s PDF, also known as Version of Record (includes final page, issue and volume numbers)

Please check the document version of this publication:

• A submitted manuscript is the version of the article upon submission and before peer-review. There can be important differences between the submitted version and the official published version of record. People interested in the research are advised to contact the author for the final version of the publication, or visit the DOI to the publisher's website.

• The final author version and the galley proof are versions of the publication after peer review.

• The final published version features the final layout of the paper including the volume, issue and page numbers.

Link to publication

General rights

Copyright and moral rights for the publications made accessible in the public portal are retained by the authors and/or other copyright owners and it is a condition of accessing publications that users recognise and abide by the legal requirements associated with these rights. • Users may download and print one copy of any publication from the public portal for the purpose of private study or research. • You may not further distribute the material or use it for any profit-making activity or commercial gain

• You may freely distribute the URL identifying the publication in the public portal.

If the publication is distributed under the terms of Article 25fa of the Dutch Copyright Act, indicated by the “Taverne” license above, please follow below link for the End User Agreement:

www.tue.nl/taverne Take down policy

If you believe that this document breaches copyright please contact us at: openaccess@tue.nl

providing details and we will investigate your claim.

(2)

An

Off

-Line Two-Dimensional Analytical Procedure for

Determination

of

Polycyclic Aromatic Hydrocarbons in

Smoke Aerosol

H.

A.

Claessens*

Eindhoven University of Technology, Lab. Instrumental Analysis, Dept. Chemical Engineering, P.O. Box 51 3, 5600 M B Eindhoven, The Netherlands

L. G. D. Lammerts van Bueren

Chrornpack International, R & D Department, P.O. Box 8033,4330 EA Middelburg, The Netherlands

Key Words:

Liquid chromatography, HPLC Two-dimensional HPLC Smoke aerosol

Polycyclic aromatic hydrocarbons Stoves

Summary

Smoke aerosol from stoves consists of a wide variety of chemical substances of which a number have toxic properties. To study the impact of aerosol emissions on health and environment reliable analytical procedures must beavailable for these samples. An off-line two-dimensional HPLC method is described for the determination of a number of polycyclic aromatic hydrocarbons (PAH). The method consists of a HPLC clean-up step yielding distinct fractions on activated silica and followed by analysis of each of the fractions by isocratic reversed phase HPLC. Detection is by UV and fluorescence monitors in series. Combination of the chromatographic data obtained from both the clean-up and analytical step provides additional qualitative information.

1 Introduction

The growing use of household stoves burning fuels such as coal, wood, etc. is resulting in increased emissions of toxic compounds. The smoke emitted from these stoves is extremely complex as it consists of a gaseous part and a solid aerosol part. Its composition is an intricate function of type of stove, fuel, and operating conditions. The major part of the toxic smoke components does not occur freely in combustion samples, but is up to 90% associated with aerosol particles in the respirable range [1,2]. Polycyclic aromatic compounds (PAC), of which a number have mutagenic and carcinogenic properties [3-51, constitute an important group of toxic components in smoke aerosol. The study of the impact

of

aerosol emission from stoves on the environment and health, and the evaluation of new stove designs, has created an urgent need for analytical procedures for the determination of smoke content. Owing to their highly toxic properties, the development of extraction and analytical procedures forthe well-known 16 Priority Pollutant PAH in smoke aerosol samples is of

primary interest. For complex samples, analytical procedures generally consist of extraction, clean-up, and preconcentration steps, followed by the final analytical stage.

Significant losses of PAH are reported by (photo)chemical decomposition and/or evaporation during storage of aerosol-loaded filters and liquid extracts [7,8]. To avoid such effects stringent precautions must be strictly observed.

To separate the organic from the inorganic part of the sample Sohxlet and/or ultrasonic extraction procedures are usually applied, using solvents such as dichloromethane, acetone, toluene, diethyl ether, cyclohexane, or mixtures thereof [8-lo]. The efficiency of the various extraction procedures depends on the chemical and physi- cal composition of the sample, the properties of the extraction solvent and the extraction conditions [lo-1 21.

Owing to the relatively large amount of carbonaceous material in this type of aerosol sample, extraction efficien- cies may be drastically reduced [13,14].

Due to the limited separation power of HPLC systems, complex samples cannot be separated completely in one step. Therefore, HPLC analysis commonly has to be preceded by a clean-up step, consisting of liquid-liquid or liquid-solid extractions or HPLC techniques [15-181.

In this paper we present a HPLCclean-up procedure, based on the silica activation procedure

of

Bredeweg [19], which allows a reproducible PAH prefractionation of the sample. Moreover, this procedure provides additional qualitative information.

Preconcentration of the sample solution is often necessary to meet the detectability of the analytical system. In order to increase the component concentrations, solvents are

342

Journal of High Resolution Chromatography & Chromatography Communications 0 1987 Dr. Alfred Huethig Publishers

(3)

PAH in Smoke Aerosol

often removed, e.g. by evaporation, which may cause losses in a number of cases. The addition of a high boiling liquid prior to this step to limit losses of PAH was investigated.

A number of HPLC systems such as normal-phase [20,21], reversed-phase [22,23], and liquid-liquid 124,251 has been developed for the separation of PAH components in both isocratic and gradient modes. Here we use an isocratic reversed-phase HPLC system under thermostated conditions, allowing automated qualitative and quantitative evaluations.

All PAH absorb in the UVspectral region, but only some are fluorescent thus permitting selective detection. Therefore, fluorescence and UV-detectors are used in series to monitor the eluent. In some cases, the ratio of UV- and fluorescence detector response provided additional qualitative information. To avoid quenching of the fluorescence signal and preferential evaporation of the volatile component in the aqueous-organic eluent, an intermittent degassing procedure proved to be more suited than the generally used continuous degassing method.

As predicted by theory, the detection limit of PAH was further improved by the use of small bore columns. Three discrimination criteria were considered to identify PAH components in the sample:

i) presence of PAH in distinct fractions in the clean-up step

ii) k‘ values of peaks in the chromatogram of the analysis step

iii) fluorescencelUV signal ratios of the peaks, showing distinct values for PAH components.

Quantitative information was obtained from peak area measurements.

2 Experimental

A number of stoves burning hardwood as a fuel were operated under well defined conditions. Aerosol samples

Table 1

Thefiguresindicate thesequenceof elutionof thePAH underthe experimental conditions as in Fig. 3.

-

Journal of High Resolution Chromatography & Chromatography Communications VOL. 10, JUNE 1987

343

No. Component No. Component

Naphthalene Acenaphth ylene Acenaphthene Fluorene Phenanthrene Anthracene Fluoranthene Pyrene 9 10 11 12 13 14 15 16 Benzotalanthracene Chrysene Benzo[b]fluoranthene Benzo[k]fluoranthene Benzo[a]pyrene Dibenzo[a,h]anthracene Benzo[ghi]perylene Indeno[l,2,3,-c,dIpyrene

were taken by the filter collection technique [6] on glass fiber filters (High Volume Airsampler, 8” X

lo”,

Pleuger, Den Bosch, The Netherlands).

2.1 Solvents and Chemicals

All chemicals and solvents were of analytical grade (Merck, Darmstadt, FRG) except petroleum ether (Shell, Rotter- dam, The Netherlands). Standards of the 16 Priority Pollutant PAH were from Serco Inc., Roseville, MI USA,

Table 1.

2.2 Chromatographic Equipment

HPLC equipment for the prefractionation and analytical steps was composed of parts of several manufacturers. The clean-up equipment consisted of a Waters pump (Model 6000A, Waters Associates, Milford, MA USA), a Valco injection valve equipped with a 200 pI loop (model CV-6-UHPa-NG0, Valco, Houston, TX USA), and a variable wavelength UV-detector (type LC-3, Pye Unicam, England) operated at 254 nm.

The home-made columns for the clean-up step, 100 X 4 mm i.d. (Knauer, Bad Homburg, FRG) were packed with 5 pm Lichrosorb Si-60 silica (Merck, Darmstadt, FRG). The analytical equipment consisted of a Beckman pump (Model 100-A, Beckman Instruments Inc., Berkeley, CA USA), an automated injection system (model 71 06, Waters Associates, Milford, MA USA), a variable wavelength UV- detector (type LC-3, Pye Unicam, England) operated at 254 nm, a fluorescence detector (model FS 970C, Kratos, Bedford, MA USA) operated at excitation and emission wavelengths of 280 and >389 nm respectively. Columns 250X 4.6mmand200X 3.0mm i,d.,packedwithVydac201 TP as stationary phase, were purchased from Chrompack (Middelburg, The Netherlands).

The analytical system, eluent bottle as well as the column were thermostated at 25OC by a circulating thermostat (Haake, Berlin) FRG.

Peaks were integrated using a Spectra Physics 4000 system using two ADC interfaces model 2 SP-4020 (Spectra Physics, Santa Clara, CA USA).

2.3 Extraction

A representative part of 2.5 cm2 of the filters was ultra- sonically extracted with four batches of 3

ml

diethyl ether each. The quicker ultrasonic extraction was preferred over the lengthy Sohxlet extraction. The four extracts were combined and evaporated to dryness under a gentle stream of N2 at ambient temperature. Samples as well as extracts were kept in sealed vessels and stored in the dark at -1 6OC. The samples were analyzed within 5 days.

(4)

2.4 Clean-Up

The extract residue was dissolved in 500pln-hexane. 250pl

Of this solution was to a HPLC clean-up

procedure on activated silica.

A closed loop system (Figure 1) was developed which

allowed reproducible clean-up and separation of the 16 priority PAH into six distinct fractions (Figure 2)). In closed loop systems the water content of the silica, which strongly influences the separation properties, can be controlled properly. n-Hexane and dichloromethane were distilled and dried over molecular sieves (3A) before use.

Figure 1

Closed-loop sample clean-up system: 1 = argon; 2 = n-hexane; 3 =

dichloromethane; 4 = activating mixture according to Bredeweg; 5 =

selecting valve; 6 = pump; 7 = injection valve; 8 = silica column; 9 = UV-

detector; 10 = three-way valve.

F r a c t i o n s I 1 ; 2 3 I 4 5 ; 6 1

time

-

I 5 min.

No. Component Fraction

1 2 3 4 5 6 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 Naphthalene Acenapthylene Acenaphthene Fluorene Fenanthrene Anthracene Fluoranthene Qrene Benz[a]anthracene Chrysene Benzo[b]fluoranthene Benzo[k]fluoranthene Benzo[a]pyrene Di benzo[a, hlant hracene Benzob, hJperylene Indeno[l,2,3-c,apyrene X X X X X X X X X X X X X X X X

The in-situ activation of the silica column, removing all water from it, is performed chemically with dimethoxy- propane, according to the procedure of Bredeweg. After that, dry n-hexane is pumped as eluent through the column and the system

is

prepared forthe clean-up procedure.The activated silica in the column will trap traces of water from the eluent and the injected samples. Depending on the state of deactivation of the silica the activation procedure must be repeated from time to time. An argon atmosphere was kept in the eluent bottle to avoid penetrating of atmospheric moisture.

The six fractions of about 2-4 ml each were collected separately and after addition of 250 pl ethyleneglycol, n-hexane was carefully evaporated in a gentle stream of N2 at ambient temperature. The results of the addition of ethylene glycol to the n-hexane fractions prior to evaporation are presented in Table 2. The time schedule to

collect the fractions from sample injections was determined with the standard PAH mixture.

Table 2

Recoveries of standard PAH amounts after evaporation in a gentle stream of N2 at ambient temperature of A = 6 ml n-hexane; B = 6 ml n-hexane after addition of 250 pI ethylene glycol; concentration of the different PAH in the n-hexane solutbns varied from 0.08-0.4 pg/ml; o = standard deviation; number of experiments, n = 6. No. Component A 0 B 0 1 2 5 6 7 8 9 10 11 12 13 14 15 16 Naphthalene Acenaphthylene Phenanthrene Anthracene Fluoranthene Pyrene Benzatalanthracene Chrysene Benzo[b]fluoranthene Benzo[k]fluoranthene Benzo[a] pyrene Dibenzo[a,h]anthracene Benzo[g,h,f]perylene Indeno[l,2,3-c,apyrene 0 0 80 80 93 91 95 91 95 92 103 94 111 98 - 5 5 4 6 4 7 4 3 - 2 2 6 9 33 83 96 94 97 97 94 94 96 96 95 97 93 99 8 9 7 7 2 3 3 9 2 2 2 6 4 8 2.5 Analysis

10 p1 of each fraction was injected and analyzed on the Vydac column. Two different columns, internal diameters 4.6 and 3.0 mm, were tested. A mixture of water and

Figure 2

Chromatogram of clean-up step of the 16 parent PAH under investiga-

tion; column, Lichrosorb Si-60,5 pm, 100 X 4.0 mm; eluent, anhydrous

n-hexane; flow rate 1.2 ml/min at ambient temperature.

(5)

I

0 9 1 1 0 10 12

I

11 Figure 3

lsocratic chromatogram of a testmix of the 16 PAH at 25OC; peak

numbers correspond to PAH nos. in Table 1.

Column, Vydac 201 TP, 5 pn, 200 X 3.0 mm; eluent, acetonitrilelwater

13 _{78:22 v:v; flow rate 0.7 mllmin.}

14

16

A Fluorescence signal

5 10

Table 3

k' valuesof PAHat 20and 25°C;column,Vydac201 TP,5pm,200

X 3.0 mm; eluent acetonitrile/water 7822 v/v, flow rate 0.7 ml/ rnin; k' values are calculated from 20 experiments, relative standard deviation of the time measurements was <0.2% for each component. 100% No. Component k'l k ' 2 k ' l - k ' d (2OOC) (25OC) k ' 2 , 1 2 314 5 6 7 8 9 10 11 12 13 14 15 16 Naphthalene Acenaphthylene Acenaphthene/fluorene Phenanthrene Ant h race ne Fluoranthene Pyrene Benzatalanthracene Chrysene Benzo[b]fluoranthene Benzo[k]fluoranthene Benzo[a] pyrene Dibenzo[a,h]anthracene Benzo[g,h,flperylene lndeno [ 1,2,3-c,q pyrene 1.13 1.27 1.60 . 1.93 2.33 2.77 3.28 5.47 6.36 10.00 12.94 14.98 23.29 23.29 28.94 1.10 1.23 1.52 1.79 2.10 2.46 2.90 4.48 5.1 1 7.72 9.59 1 1.30 16.36 17.47 20.57 2.7 3.3 5.3 7.8 11.0 12.6 13.1 22.1 24.5 29.5 34.9 32.6 42.4 33.3 40.7 15 2 0 __3 Time (min)

acetonitrile was used as an eluent. The influence of the temperature on resolution and k' values was investigated

between 20 and 35OC. An example of the influence of the temperature on k' values of PAH under standard conditions is presented in Table 3. Figure 3 shows a chromatogram of

a standard mixture of 16 PAH on a 3.0 mm i.d. column. A short degassing procedure of 5 min per 24 h with argon was applied, after which the eluent bottle was sealed. To avoid pump problems arising from underpressure and for reasons of economy, the eluent was recirculated. The reproducibility of the quantitative analysis of the PAH in the analytical step was studied; the results are presented in

Table 4.

3 Results and Discussion

The closed loop clean-up method offers a system for the separation of the 16 PAH and other sample components, which might coelute under these conditions, in six distinct peaks. In practice, the clean-up system proved to be highly reliable and reproducible. Hence retention data of this step provided additional qualitative information of the samples under investigation. Therefore, in the final analytical stage coeluting interfering sample components can be discrimi-

(6)

Table 4

Reproducibility of the PAH measurements in the analytical step. Column: Vydac 201 TP, 5 pm, 200 X 3.0 mm; eluent: acetonitrile/ water 78:22 v/v, flow rate 0.7 ml/min; temperature: 25OC. The data of the components 1 to 6 are calculated from the UV detection signal; the data of the other components are calculated from the fluorescence detection signal; number of experiments, n = 5; m = injected amount of PAH; V = relative standard deviation.

No. Component m (ng) V (YO) Detection limit (ng) 1 Naphthalene 45.0 6 0.13 2 Acenaphthylene 38.2 5 3.0 3 Acenaphthene 4 Fluorene coelution 42.0 4 9.8 5 Phenanthrene 10.1 3 0.09 6 Anthracene 6.6 2 0.04 7 Fluoranthene 20.2 5 0.03 8 Pyrene 19.7 13 0.09 9 Benza[a]anthracene 10.1 2 0.01 11 Benzo[b]fluoranthene 10.2 2 0.01 12 Benzo[k]fluoranthene 10.0 2 0.01 13 Benzo[a]pyrene 10.6 3 0.01 14 Dibenzo[a,h]anthracene 7.4 5 0.02 15 Benzo[g,h,flperylene 8.0 3 0.02 10 Chrysene 9.4 4 0.13 16 Indeno[1,2,3-c,qpyrene 8.1 4 0.04

nated. A drawbackof this clean-up step is that each of the six collected fractions has to be analyzed for all

of

the 16 PAH, thus increasing total analysis time. The regenera- tion frequency of the clean-up system depends on quantity and composition of the injected samples.

The influence

of

the addition of a high boiling liquid such as ethylene glycol prior to the evaporation of n-hexane was investigated by evaporation of thestandard solution of PAH inn-hexane.The resultsfromTable2showthat additionofa high boiling liquid prior to evaporation of n-hexane improves PAH recoveries substantially.

An isocratic eluent mixture of waterlacetonitrile 22:78 v/v showed optimal resolution with the Vydac columns. The columns of 3.0 mm i.d. gave better signal-to-noise ratios, which are a function

of

the square of column radius. These columns also showed better peak shapes, allowing more accurate computer calculations with respect to the recognition of peak contours. Moreover, under otherwise identical chromatographic conditions, smaller inside dia- meter columns can be operated at lower eluent flows. PAH capacity factors depend strongly on the temperature of the column, as can be concluded e.g. from the results presented in Table 3, where the increase of k' values over 5OC has been calculated. Automatic peak identification requires fairly constant retention times and hence thermo- stating of the column. A temperature of 25°C resulted in optimal peak resolution for this particular HPLC system.

Degassing of the eluent is necessary to avoid quenching of the fluorescence signal. Continuous degassing by helium causes preferential evaporation of acetonitrile from the aqueous-organic eluent, resulting in shifts of retention times of chromatographic peaks. The argon degassing procedure applied here combined with the thermostated HPLC analytical system offered highly reproducible k'

values (Table 3) over a number of weeks. Moreover for reasons of economy, it is attractive to use the same batch of eluent for 4-6 weeks before preparing a new one. To avoid premature increase of drift of the baseline the eluent was homogenized once a day.

Identification of PAH was based on:

i) presence in a particular peak in the clean-up step ii) retention times of the peaks in the sample compared to

standard solutions in the analytical step iii) the ratio UV to fluorescence detector signal.

In Figure 4 the qualitative results, based on the retention behavior of the PAH of the two-dimensional chromatographic separations, are summarized. From these results it can be concluded that combination of the straight-phase chromatography for sample clean-up procedure and reversed phase analytical steps provides the possibility

of

complete separation of the 16 PAH.

21

m Y 20 1 15 10 5

.

I 6

.>..

.

10 . 9 .i - 8 . b .s

.'

;;

I 1 , I , , , , X I I"

,

Y

,

-

1 2 3 4 5 6 7 8 time (minutes) Figure 4

Plot of the chromatographic behavior of the PAH in the off-line clean-up and separation steps; Xaxis, timescheduleof theclean-upstepoverthe

six fractions (Roman numerals); Y axis, k' values of the PAH under

standard chromatographic conditions; numberscorrespond to PAH nos. in the text.

(7)

The PAH amounts in the aerosol samples were calculated from peak area measurements, which were compared to the corresponding values of standard PAH injections. The reproducibility of the quantitative analysis step proved to be satisfactory, as can be concluded from Table 4, furthermore, the detection limits for 14 of the 16 PAH are calculated.

The relatively high detection limit of acenaphthylene might be caused by two factors:

i) Acenaphthylene has the lowest molar absorbance at 254 nm of the PAH under study;

ii) the acenaphthylene peak has been integratedasa rider peak on naphthalene.

4 Conclusions

Aselective clean-up procedure forthe 1 6priority pollutants PAH in aerosol samples is presented. This procedure provides additional qualitative information to the information of the analytical step. The analytical system, operating overnight unattended, proved to give reliable qualitative and quantitative results.

Acknowledgments

We thank Mr. H. J. Ritchie and Mr, G. A. F. M. Rutten for revising the manuscript and Mrs. D. C. M. Tjallema for her technical assistance.

References [l]

[2]

[3]

F. DeWiest, Atmos. Environ. 12 (1978) 1705.

G. Broddin, L. van Vaeck, and K. van Cauwenberghe, Atmos. Environ. 11 (1 9777) 1061.

M. L. Lee, M. V. Novotny, K. D. Bartle, “Analytical Chemistry of Polycyclic Aromatic Hydrocarbons”, Academic Press, New York (1981).

[4] A. Bjorseth (Ed.), “Handbook of Polycyclic Aromatic Hydro- carbons”, Marcel Dekker, New York (1983).

[5] fh. Ramdahl, “Determination of Polycyclic Aromatic Compounds (PAC) in Source Emission and Ambient Air”, Thesis, Oslo (1 984).

H. Zeedijk, J. Aerosol Sci. 17 (1 986) 635.

Science 202 (1978) 515.

P. J. Robertson, R. H. Groth, D. G. Gardner, et. a/., J. Air Pollut. Control Assoc. 29 (1979) 143.

G. Chatot, J. Chromatogr. 72 (1972) 202.

Pollut. 6 (1976) 103.

[ l l ] W. F. Fitch and D. H. Smith, Environ. Sci. Technol. 13 (1979)

341.

[12] W. H. Griest, L. B. Yeatts, and J . E. Caton, Anal. Chem. 52

(1980) 199.

[13] A. Tateda and J . W. Fritz, J. Chromatogr. 152 (1978) 329. [14] H. A. Claessens, L. G. D. Lammerts van Bueren, and P. M. van

de Ven, J. Aerosol Sci. 17 (1986) 639.

[15] B. P. Dunn, R. J . Armour, Anal. Chem. 52 (1980) 2027. [16] W. J. Sonnefeld, W. H. Zoller, W. E. May, et. a/., Anal. Chem.

54 (1 982) 723.

[17] T. Spitsen, J. Chromatogr. 237 (1982) 273.

[18] H. Stray, S. Mano, A. Nikolsen et. a/., HRC & CC 7 (1984) 74. [19] R. A. Bredeweg, L. D. Rothman, C. D. Pfeiffer, Anal. Chern. 51

(1 979) 2061.

[20] T. V. Alfredson, J. Chromatogr. 218 (1981) 715.

[21] H. Menard, L. Noel, F. H. Kimmerle, et. a/., Anal. Chem. 56

(1 984) 1 240.

1221 M. N. Hasan, P. C. Jun, Anal. Chem. 55 (1983) 263. [23] R. J. Hurtubise, T. W. Allen, H. F. Silver, J. Chromatogr. 235

[24] K. P. Hupe, H. Schrenker, Chromatographia 13 (1 980) 350. [25] J. P. Crombeen, S. Heemstra, and J . C. Kraak, J. Chromatogr.

Ms received: April 8, 1986

Accepted by WB: March 10,1987 [6]

[7] J. N. Pitts, K. van Cauwenberghe, D. Grosjean, et. a/.,

[8]

[9]

[lo] W. Cautreels and K. van Cauwenberghe, Water Air Soil

(1 982) 51 7.

286 (1 984) 1 19.