Chemosphere 73 (2008) 614–618

Contents lists available at ScienceDirect

Chemosphere
j o u r n a l h o m e p a g e : w w w . e l s e v i e r. c o m / l o c a t e / ch e m o s p h e r e

Identification of stationary sources of air pollutants by concentration statistical

analysis
Pasquale Iovino, Stefano Salvestrini, Sante Capasso *
Depart­ment of Envi­ron­men­tal Sci­ences, Sec­ond Uni­ver­sity of Naples, via Vivaldi 43, Ca­ser­ta, Italy

a r t i c l e i n f o a b s t r a c t

Article history: The atmo­spheric con­cen­tra­tions of ben­zene, tol­u­ene, eth­yl­ben­zene and iso­meric xylenes (BTEX) in a
Received 13 Feburary 2008 medium-sized town (S. Maria Cap­ua Ve­tere, about 32000 inhab­it­ants, South­ern Italy) have been deter­
Received in revised form 20 May 2008 mined dur­ing work­ing days and week­ends in 2006. The pro­ce­dure used was 24 h pas­sive adsorp­tion by
Accepted 21 May 2008
sam­plers dis­trib­uted through­out the town fol­lowed by GC/MS anal­y­sis. On a yearly base, the arith­me­tic
Available online 21 July 2008

mean ben­zene con­cen­tra­tions were above the limit required by the 2000/69/CE Euro­pean Direc­tive. The
Pear­son cor­re­la­tion coef­fi­cients of the 24 h geo­met­ric mean BTEX con­cen­tra­tions were indic­a­tive of sta­
tion­ary sources of tol­u­ene located in a well cir­cum­scribed area of the urban ter­ri­tory, active only dur­ing
Key­words:
BTEX
the work­ing days and not offi­cially rec­og­nized. The results high­light the effec­tive­ness of the sta­tis­ti­cal
Atmo­spheric pol­lu­tion approach used in this study for the iden­ti­fi­ca­tion of pol­lu­tant sources.
Vehic­u­lar traf­fic © 2008 Else­vier Ltd. All rights reserved.
Sta­tion­ary sources
Dif­fu­sive sam­plers

1. Intro­duc­tion
Aro­matic hydro­car­bons, in par­tic­u­lar ben­zene, tol­u­ene, eth­
yl­ben­zene and xylenes (BTEX) are among the main pol­lu­tants in
urban areas (Sein­feld and Pan­dis, 1998) and their adverse effects
on human health are well doc­u­mented (Vic­to­rin, 1993). Sev­eral
stud­ies (Hin­wood et al., 2006; Hin­wood et al., 2007; Ker­ba­chi et
al., 2006; Mar­tins et al., 2007; Yas­saa et al., 2006) have shown that
the major con­tri­bu­tion to BTEX in urban areas is from the vehic­u­lar
traf­fic. Organic com­pounds emit­ted from the exhaust pipe con­sist
mainly of unburned aro­matic fuel com­po­nents (Siegl et al., 1992).
The air con­tent of ben­zene rel­a­tive to other aro­matic com­pounds
is sig­nif­i­cantly higher than its fuel per­cent­age because ben­zene is a
par­tially oxi­dized prod­uct of other aro­matic com­pounds (McGinty
and Dent, 1995). The air con­cen­tra­tion of ben­zene is sub­ject to
restric­tion in many coun­tries whilst for the other BTEX the norms
are usu­ally more per­mis­sive. The Direc­tive 2000/69/CE of the Euro­
pean Union has fixed the limit ben­zene con­cen­tra­tion to 5 lg m¡3
for the urban atmo­sphere.
Besides vehic­u­lar traf­fic, emis­sion from sta­tion­ary sources,
mainly fac­to­ries, also plays a rel­e­vant role in affect­ing the atmo­
sphere qual­ity (Shar­ma et al., 2007; Sri­vast­av­a, 2004). Vol­a­tile
organic com­pounds (VOCs) includ­ing BTEX are widely used in
indus­tries and their emis­sion in the atmo­sphere is a con­stant risk
for the pop­u­la­tion. Fuel sta­tions, small fac­to­ries, main­te­nance
* Cor­re­spond­ing author. Tel.: +39 082 3274623; fax: +39 082 3274605.
E-mail address: San­te.Cap­as­so@un­in­a2.it (S. Capasso).
shops of house-hold appli­ances and garages are the main urban
sta­tion­ary sources of VOCs. The assess­ment of vehic­u­lar and non-
vehic­u­lar con­tri­bu­tion to VOCs is an indis­pens­able step for the
enforce­ment of mea­sures directed at improv­ing the qual­ity of
urban air. Cer­tain VOCs assumed to be emit­ted exclu­sively from
vehi­cles have been used as indi­ca­tors of vehic­u­lar pol­lu­tion, for
exam­ple methyl tert-butyl ether, an octane-enhanc­ing gas­o­line
addi­tive (Chang et al., 2006). This approach, how­ever, should be
used with cau­tion, as the rel­a­tive con­tents of BTEX com­pounds in
the atmo­sphere change with the aging of pol­lu­tants, because of
their dif­fer­ent deg­ra­da­tion rate (Brocco et al., 1997). Other fac­tors
affect­ing VOCs con­cen­tra­tions include the car speed and nota­bly
the dif­fu­sion and effi­ciency of cat­a­lytic con­vert­ers (Heeb et al.,
2000).
In the pres­ent study vehic­u­lar and sta­tion­ary emis­sions in a
medium-sized town in South­ern Italy were assessed by a novel
pro­ce­dure based on the sta­tis­ti­cal anal­y­sis of spa­tial and tem­po­ral
vari­a­tions of BTEX con­cen­tra­tions.
2. Exper­i­men­tal
2.1. Sam­pling sites
The BTEX con­cen­tra­tions have been deter­mined in the atmo­
sphere of S. Maria Cap­ua Ve­tere, about 32000 inhab­it­ants, in South­
ern Italy (lat. 41° 59 N, long. 14° 159 E).
The loca­tion of the sam­pling sites is indi­cated in Fig. 1. The sam­
pling was per­formed in work­ing days and on week­ends through­out
2006, except August. We per­formed 1 or 2 sam­plings per month

0045-6535/$ - see front matter © 2008 Else­vier Ltd. All rights reserved.
doi:10.1016/j.chemosphere.2008.05.047
 P. Iov­i­no et al. / Chemosphere 73 (2008) 614–618 615
Fig. 1. Map of the sam­pling points in S. Maria Cap­ua Ve­tere.
in work­ing days and six sam­plings on week­ends for the whole
year, from Sat­ur­day even­ing to Sun­day even­ing. The data rel­a­tive
to rainy days were not included. All sam­pling points were located
along the streets of the city, about 5 m above the ground level and
not in prox­im­ity of cross-roads.
2.2. Ana­lyt­i­cal pro­ce­dures
The BTEX com­pounds (ben­zene, tol­u­ene, eth­yl­ben­zene, o-
xylene and (m + p)-xylene) were col­lected by Rad­i­ello® dif­fu­sive
sam­plers, pat­ented by Foun­da­tion Sal­va­to­re Mau­ge­ri (FSM). The
expo­si­tion time for each mon­i­tor­ing cam­paign was 24 h. The sam­
plers con­sisted of a stain­less steel net cyl­in­der with a 100-mesh-
grid open­ing, packed with 530 mg of acti­vated char­coal. The anal­
y­ses were car­ried on a Perkin Elmer GC–MS instru­ment, model
Cla­rus 500, equipped with a ther­mal desorp­tion instru­ment,
model Tur­boM­a­trix ATD. The ana­lytes, extracted from exposed car­
tridges by a stream of helium (80 ml min¡1 at 320 °C for 10 min),
were condensed in a trap at 2 °C and vapor­ized by heat­ing at 290 °C
for 1 min.
The col­umn used was an Elite-5MS fused sil­ica cap­il­lary,
30 m £ 0.25 mm, 0.25 lm film thick­ness (Perkin Elmer Instru­
ments). The oven tem­per­a­ture pro­gram was 35 °C for 5 min, to
120 °C at 8 °C min¡1 and to 200 °C at 15 °C min¡1. The car­rier gas
was He at a flow rate of 1 ml min¡1. The mass spec­trom­e­ter scanned
from 35 to 300 a.m.u. every 0.1 s, in the elec­tronic impact (70 eV)
mode. The ion source tem­per­a­ture was 180 °C and the mul­ti­plier
volt­age 350 V.
2.3. Sta­tis­ti­cal anal­y­sis
Sta­tis­ti­cal anal­y­sis was car­ried out using the sta­tis­ti­cal soft­ware
pack­age Stat­is­ti­ca 7 of Stat­Soft.
3. Results and dis­cus­sion
3.1. Con­cen­tra­tions in the sam­pling sites
Although this sur­vey was car­ried out in a medium-sized town,
the geo­met­ric mean con­cen­tra­tions were rel­a­tively high with
616 P. Iov­i­no et al. / Chemosphere 73 (2008) 614–618

Table 1
Geo­met­ric mean BTEX con­cen­tra­tions (lg m¡3) recorded in 16 sam­pling sites on work­ing days and week­ends in S. Maria Cap­ua Ve­tere (South­ern Italy) dur­ing 2006; geo­met­
ric stan­dard devi­a­tions (%) in paren­the­sis

Site Day Ben­zene Tol­u­ene Eth­yl­ben­zene (m + p)-xylene o-xylene

1 Work­ing days 3.6 (27) 10.7 1.9 (25) 4.3 (24) 2.0 (21)
Week­ends n.d. n.d. n.d. n.d. n.d.
2 Work­ing days 4.5 (23) 14.9 (26) 2.9 (32) 7.0 (30) 3.4 (26)
Week­ends 5.1 (47) 10.7 (47) 2.0 (44) 7.3 (48) 2.2 (50)
3 Work­ing days 7.5 (23) 33.3 (28) 6.2 (26) 16.3 (19) 8.0 (29)
Week­ends 5.6 (18) 12.0 (22) 2.6 (15) 7.3 (5) 2.8 (10)
4 Work­ing days 5.3 (21) 14.0 (29) 3.1 (31) 7.9 (29) 4.0 (31)
Week­ends 4.3 (13) 8.2 (22) 2.0 (28) 6.2 (20) 2.4 (23)
5 Work­ing days 10.3 (22) 31.5 (24) 6.9 (25) 17.9 (28) 9.4 (26)
Week­ends 7.9 (27) 19.5 (23) 4.1 (21) 10.9 (27) 4.4 (24)
6 Work­ing days 3.8 (24) 11.9 (29) 2.4 (30) 5.8 (28) 2.9 (33)
Week­ends 3.3 (40) 11.2 (43) 2.4 (43) 7.2 (41) 2.5 (44)
7 Work­ing days 5.3 (24) 18.1 (26) 3.5 (31) 8.8 (30) 4.3 (29)
Week­ends 6.1 (27) 11.7 (30) 3.0 (27) 8.7 (28) 3.5 (27)
8 Work­ing days 4.8 (25) 16.1 (30) 3.1 (31) 7.9 (30) 3.9 (23)
Week­ends 3.4 (21) 9.5 (4) 2.0 (22) 7.0 (22) 2.7 (29)
9 Work­ing days 4.2 (33) 22.2 (28) 2.9 (30) 7.4 (21) 3.5 (26)
Week­ends n.d. n.d. n.d. n.d. n.d.
10 Work­ing days 6.1 (29) 19.4 (33) 4.3 (28) 10.6 (30) 5.9 (32)
Week­ends 5.7 (23) 13.1 (18) 3.5 (31) 9.6 (19) 3.8 (35)
11 Work­ing days 8.3 (21) 24.6 (24) 4.5 (30) 9.3 (38) 4.9 (35)
Week­ends 5.8 (14) 14.9 (34) 3.0 (25) 9.4 (21) 3.6 (30)
12 Work­ing days 4.5 (28) 14.4 (29) 3.1 (33) 8.1 (35) 4.0 (32)
Week­ends n.d. n.d. n.d. n.d. n.d.
13 Work­ing days 7.0 (31) 22.5 (26) 4.4 (34) 10.8 (32) 5.4 (34)
Week­ends 7.0 (38) 16.7 (40) 3.5 (44) 10.8 (42) 4.0 (43)
14 Work­ing days 10.2 (23) 33.8 (23) 7.4 (27) 19.2 (27) 10.1 (32)
Week­ends 5.1 (29) 11.3 (27) 2.4 (25) 7.9 (34) 3.0 (40)
15 Work­ing days 6.1 (35) 17.8 (33) 3.9 (33) 9.5 (32) 5.0 (28)
Week­ends n.d. n.d. n.d. n.d. n.d.
16 Work­ing days 7.1 (24) 24.7 (32) 4.2 (31) 10.1 (35) 5.1 (30)
Week­ends 7.2 (25) 23.4 (39) 4.3 (33) 11.4 (27) 4.3 (31)
All sites Work­ing days 5.9 (35) 19.4 (29) 3.8 (25) 9.3 (23) 4.7 (34)
Week­ends 5.4 (31) 12.8 (28) 2.8 (21) 8.4 (22) 3.1 (32)
n.d. = not deter­mined.

r­ ef­er­ence to the limit val­ues required by the Euro­pean Union. For
exam­ple, we recorded a geo­met­ric mean value for ben­zene, com­
puted using data from all the sites, of 5.9 lg m¡3 and of 5.4 lg m¡3
for the work­ing days and week­ends, respec­tively. Both these val­ues
are higher than the limit value of 5 lg m¡3 fixed by the EU. Table 1
sum­ma­rizes the geo­met­ric mean BTEX con­cen­tra­tions recorded in
the mon­i­tored town dur­ing 2006. As can be seen, there is a marked
dif­fer­ence from site to site, prob­a­bly reflect­ing dif­fer­ences in the
traf­fic den­sity and in the street topog­ra­phies near the sam­pling
sites. The geo­met­ric mean are char­ac­ter­ised by rel­a­tively high geo­
met­ric stan­dard devi­a­tions, most likely reflect­ing dif­fer­ent mete­o­
ro­log­i­cal con­di­tions dur­ing the sam­pling days.
Fig. 2 reports the geo­met­ric mean BTEX con­cen­tra­tions recorded
on work­ing days and on week­ends. Except for tol­u­ene, the con­cen­
tra­tion of each BTEX com­pound was slightly lower on week­ends,
prob­a­bly reflect­ing a reduc­tion in the vehic­u­lar traf­fic den­sity. Tol­u­
ene con­cen­tra­tions were mark­edly higher in work­ing days, sug­gest­
ing a dif­fer­ent dynam­ics.
3.2. Sta­tis­ti­cal anal­y­sis
Table 2 reports the lin­ear cor­re­la­tion matri­ces of Pear­son for
the 24 h geo­met­ric mean BTEX con­cen­tra­tions recorded on work­
ing days and week­ends. For both matri­ces the coef­fi­cients are
close to one, indi­cat­ing that the BTEX con­cen­tra­tions are strictly
related and most likely depend on one major source, prob­a­bly the
vehic­u­lar traf­fic. It should be noted that the aver­age car speed
and the aver­age effi­ciency of cat­a­lytic con­vert­ers may affect the
BTEX ratio, but not the cor­re­la­tion coef­fi­cients. The cor­re­la­tion
coef­fi­cients obtained in the pres­ent study are lower than those
for­merly reported by us (Iov­i­no et al., 2006) and by oth­ers (Fer­
nan­dez-Mar­ti­nez et al., 2001; Han­sen and Palm­gren, 1996) for
sim­i­lar sit­u­a­tions. It is inter­est­ing to note that the low­est coef­fi­
cients in our work were observed for tol­u­ene ver­sus all the other
pol­lu­tants. More­over, a com­par­i­son of the cor­re­la­tion matri­ces
for work­ing days, com­puted exclud­ing in turn the con­tri­bu­tion
of each site, showed that the sites 7, 9 and 11, all located in the
north of the town (Fig. 1), con­trib­uted mark­edly to lower the cor­
re­la­tion coef­fi­cients. When these were excluded from the anal­y­sis,
a marked incre­ment of the coef­fi­cient val­ues was observed. Table
3A reports the cor­re­la­tion matrix com­puted with­out the con­tri­bu­
tion of these sites.
In line with this result, the coef­fi­cient matrix com­puted using
only the data from the sites 7, 9 and 11 on work­ing days has strik­
ingly lower coef­fi­cients, par­tic­u­larly for tol­u­ene terms (Table 4A).
This is a clear indi­ca­tion that tol­u­ene was emit­ted not only from
vehic­u­lar traf­fic but also from one or sev­eral sta­tion­ary sources
close to the sites 7, 9 and 11. Com­par­ing the cor­re­la­tion matri­ces
for the con­cen­tra­tions recorded on week­ends in all sites (Table 2B),
exclud­ing the sites 7, 9 and 11 (Table 3B), or based only on the data
from sites 7, 9 and 11 (Table 4B) pro­vides addi­tional evi­dence for
the pres­ence in these sites of tol­u­ene sources other than vehic­u­
lar traf­fic. Indeed, the data col­lected on week­ends (from the Sat­ur­
day even­ing to Sun­day even­ing) showed a very good cor­re­la­tion
between all the BTEX com­pounds and no var­i­a­tion was observed
depend­ing on the sam­pling sites. It is rea­son­able to assume that
the emis­sions from sta­tion­ary source(s) were lim­ited to work­ing
days.
 P. Iov­i­no et al. / Chemosphere 73 (2008) 614–618 617

25

Working days
20
Weekends

Average concentration (µ gm- 3)

15

10

0
Benzene Toluene Ethylbenzene (m+p)-Xylene o-Xylene

Fig. 2. Geo­met­ric mean BTEX con­cen­tra­tions recorded on work­ing days and on week­ends.

Table 2 Table 4
Cor­re­la­tion coef­fi­cients between 24 h aver­age BTEX con­cen­tra­tions recorded on Cor­re­la­tion coef­fi­cients between the 24 h aver­age BTEX con­cen­tra­tion in work­ing
work­ing days (A) and on week­ends (B) days (A) and on week­ends (B) cal­cu­lated using only the data recorded in the sam­
pling sites 7, 9 and 11
Ben­zene Tol­u­ene Eth­yl­ben­zene (m + p)-Xylene o-Xylene
Ben­zene Tol­u­ene Eth­yl­ben­zene (m+p)-Xylene o-Xylene
(A)
(A)
Ben­zene 1 0.76 0.90 0.83 0.88
Tol­u­ene 1 0.81 0.71 0.77 Ben­zene 1 0.27 0.81 0.73 0.74
Tol­u­ene 1 0.46 0.37 0.36
Eth­yl­ben­zene 1 0.91 0.97
Eth­yl­ben­zene 1 0.90 0.95
(m + p)-Xylene 1 0.83
(m + p)-Xylene 1 0.82
o-Xylene 1
o-Xylene 1
(B)
Ben­zene 1 0.90 0.90 0.86 0.87 (B)
Tol­u­ene 1 0.88 0.86 0.88 Ben­zene 1 0.90 0.90 0.86 0.87
Eth­yl­ben­zene 1 0.91 0.95 Tol­u­ene 1 0.88 0.86 0.88

(m + p)-Xylene 1 0.96 Eth­yl­ben­zene 1 0.91 0.95

(m + p)-Xylene 1 0.96
o-Xylene 1
o-Xylene 1

Table 3 tol­u­ene inde­pen­dent of vehic­u­lar traf­fic active only on work­ing

Cor­re­la­tion coef­fi­cients between the 24 h aver­age BTEX con­cen­tra­tions in work­ing days and not offi­cially rec­og­nized. These results show that the anal­
days (A) and on week­ends (B) cal­cu­lated with­out the data recorded in the sam­pling
sites 7, 9 and 11
y­sis of spa­tial and tem­po­ral dis­tri­bu­tions of air pol­lu­tant con­cen­
tra­tions by means of the lin­ear cor­re­la­tion matri­ces of Pear­son is a
Ben­zene Tol­u­ene Eth­yl­ben­zene (m + p)-Xylene o-Xylene
suit­able method for iden­ti­fy­ing minor sta­tion­ary sources.
(A)
Ben­zene 1 0.90 0.91 0.84 0.90 Ref­er­ences
Tol­u­ene 1 0.92 0.82 0.90
Eth­yl­ben­zene 1 0.91 0.97 Brocco, D., Frat­arc­ang­el­li, R., Le­pore, L., Petr­ic­ca, M., Vent­rone, I., 1997. Deter­mi­na­tion
(m + p)-Xylene 1 0.84 of aro­matic hydro­car­bons in urban air of Rome. Atmos. Envi­ron. 31, 557–566.
o-Xylene 1 Chang, C.C., Wang, J.L., Liu, S.C., Lung, S.C.C., 2006. Assess­ment of vehic­u­lar and non-
vehic­u­lar con­tri­bu­tions to hydro­car­bons using exclu­sive vehic­u­lar indi­ca­tors.
(B) Atmos. Envi­ron. 40, 6349–6361.
Ben­zene 1 0.91 0.90 0.87 0.87 Fer­nan­dez-Mar­ti­nez, G., Lo­pez-Ma­hi­a, P., Mu­ni­ate­gui-Lore­nzo, S., Pra­da-Rodri­guez,
Tol­u­ene 1 0.89 0.86 0.88 D., Fer­nan­dez-Fer­nan­dez, E., 2001. Mea­sure­ment of vol­a­tile organic com­pounds
Eth­yl­ben­zene 1 0.92 0.95 in urban air of la Cor­uña Spain. Water, Air, Soil Pol­lut. 129, 267–288.
Han­sen, A.B., Palm­gren, F., 1996. VOC air pol­lu­tants in Copen­ha­gen. Sci. Total Envi­
(m + p)-Xylene 1 0.96
ron. 189 (190), 451–457.
o-Xylene 1 Heeb, N.V., Forss, A.M., Bach, C., Mat­trel, P., 2000. Veloc­ity-depen­dent emis­sion fac­
tors of ben­zene, tol­u­ene and C2-ben­zenes of a pas­sen­ger car equipped with and
with­out a reg­u­lated 3-way cat­a­lyst. Atmos. Envi­ron. 34, 1123–1137.
Hin­wood, A.L., Berko, H.N., Far­rar, D., Gal­bal­ly, I.E., Weeks, I.A., 2006. Vol­a­tile organic
4. Con­clu­sion com­pounds in selected micro-envi­ron­ments. Che­mo­sphere 63, 421–429.
Hin­wood, A.L., Rodri­guez, C., Run­nion, T., Far­rar, D., Mur­ray, F., Hor­ton, A., Glass, D.,
Shep­pe­ard, V., Edwards, J.W., Den­i­son, L., Whit­worth, T., Eis­er, C., Bul­sara, M.,
Our results pro­vide strong evi­dence for the pres­ence, in the Gil­lett, R.W., Pow­ell, J., Law­son, S., Weeks, I., Gal­bal­ly, I., 2007. Risk fac­tors for
well-defined area of the town ana­lysed, of sta­tion­ary source(s) of increased BTEX expo­sure in four Aus­tra­lian cit­ies. Che­mo­sphere 66, 533–541.
618 P. Iov­i­no et al. / Chemosphere 73 (2008) 614–618
Iov­i­no, P., Sal­ves­trin­i, S., Cap­as­so, S., 2006. Back­ground atmo­spheric lev­els of BTEX
in a medium-sized city and sur­round­ing area in South­ern Italy. In: Breb­bia C.A.
(Eds.). Air Pol­lu­tion XIV, Great Brit­ain, pp. 611–618.
Ker­ba­chi, R., Bough­e­da­oui, M., Bou­noua, L., Ked­dam, M., 2006. Ambi­ent air pol­lu­
tion by aro­matic hydro­car­bons in Algiers. Atmos. Envi­ron. 40, 3995–4003.
Mar­tins, E.M., Arb­il­la, G., Bau­er­feldt, G.F., de Paula, M., 2007. Atmo­spheric lev­els of
alde­hydes and BTEX and their rela­tion­ship with vehic­u­lar fleet changes in Rio
de Janeiro urban area. Che­mo­sphere 67, 2096–2103.
McGinty, R.P., Dent, N.P., 1995. A review of the effect of pet­rol com­po­si­tion on unreg­
u­lated motor vehi­cle emis­sions with par­tic­u­lar empha­sis on non-cat­a­lyst vehi­
cles. Envi­ron. Tech­nol. 16, 603–623.
Sein­feld, J.H., Pan­dis, S.N., 1998. Atmo­spheric chem­is­try and phys­ics. In: Wiley,
(Ed.), Air Pol­lu­tion to Cli­mate Change, New York, pp. 724–743.
Shar­ma, H., Jain, V.K., Khan, Z.H., 2007. Char­ac­ter­iza­tion and source iden­ti­fi­ca­tion
of poly­cy­clic aro­matic hydro­car­bons (PAHs) in the urban envi­ron­ment of Delhi.
Che­mo­sphere 66, 302–310.
Siegl, W.O., McC­a­be, R.W., Chun, W., Kai­ser, E.W., Perry, J., Henig, Y.I., Trin­ker, F.H., Ander­
son, R.W., 1992. Spe­ci­at­ed hydro­car­bon emis­sions from the com­bus­tion of sin­gle
com­po­nent fuels. I: effect of fuel struc­ture. J. Air Waste Man­age. 42, 912–920.
Sri­vast­av­a, A., 2004. Source appor­tion­ment of ambi­ent VOCs in Mum­bai city. Atmos.
Envi­ron. 38, 6829–6843.
Vic­to­rin, K., 1993. Health effects of urban air pol­lu­tants:guide­line val­ues and con­di­
tions in Swe­den. Che­mo­sphere 27, 1691–1706.
Yas­saa, N., Bran­cale­on­i, E., Frat­ton­i, M., Cicc­i­o­li, P., 2006. Iso­meric anal­y­sis of BTEXs
in the atmo­sphere using b-cyclo­dex­trin cap­il­lary chro­ma­tog­ra­phy cou­pled with
ther­mal desorp­tion and mass spec­trom­e­try. Che­mo­sphere 63, 502–508.