Depósi o de in es igación de la Uni e sidad de Se illa
h ps://idus.us.es/
“This is an Accep ed Manusc ip o an a icle published by Else ie in:
CHEMOSPHERE on 2006, a ailable a :
h ps://doi.o g/10.1016/j.chemosphe e.2005.06.038”
The composi ion and ela ionships be ween ace elemen le els
in inhalable a mosphe ic pa icles (PM10) and in lea es o Ne ium oleande L. and
Lan ana cama a L.
1A.J. Fe nández Espinosa, 2S. Rossini Oli a*
1 Depa men o Analy ical Chemis y, Uni e si y o Se ille, C. P o eso Ga cía González
s/n, E-41012, Se ille, Spain E.mail: [email p o ec ed]
2Depa men o Plan Biology and Ecology, Uni e si y o Se ille, A . Reina Me cedes s/n,
Apdo. 1095, 41080 Se ille, Spain, Tel.: +34954557056; Fax: +34954557059
*Co esponding au ho : e-mail: [email p o ec ed]
ABSTRACT
In o de o e alua e he in luence o ai pollu an in lux on inhalable a mosphe ic pa icles
(PM10) me al composi ion and s udy he ela ionship be ween ace elemen le els in PM10
and in lea es o wo plan species, he concen a ion o Ba, Cu, Fe, Mn, Pb, Ti and V we e
analysed in PM10 and in Ne ium oleande L. and Lan ana cama a L. lea es om 3 si es o
he ci y o Se ille. In he PM10, Cu, Fe and Mn con en was signi ican ly lowe (p0.05) in
he con ol si e han in he o he si es. A mosphe ic pollu ion le el o he ci y o Se ille can
be gene ally conside ed no dange ous o human heal h. No co ela ion be ween lea con en
and ai con en was ound o none elemen s in L. cama a. By con as posi i e and
signi ican co ela ions (p0.05) we e ound be ween lea con en o N. oleande and PM10
con en o Cu and Fe. The da a sugges ha N. oleande can be used in a mosphe ic
biomoni o ing s udies, being especially use ul o Cu and Fe and much indica e han L.
cama a.
Keywo ds: Biomoni o ing; Lea ; Pa icles; T ace elemen s
INTRODUCTION
In e es on he e ec o a mosphe ic pa icles and especially in he inhalable ac ion
(PM10) inc eased du ing he las yea s. PM10 a e conside ed o be ha m ul o human heal h
since hey can cause espi a o y and ca dio ascula disease. Associa ion be ween as hma
a acks and coughs wi h pa icula es ha e also been epo ed (F ee -Smi h e al., 1997). In
la ge u ban a eas, he conside able amoun o pa icula e ma e in he ai is a e y c ucial
p oblem. The p ima y sou ce o ai bo ne pa icula e ma e is ag icul u al p ac ises, ehicle
a ic, indus ial p ocesses and ossil uel s a ions (Becke e al., 1998). On he o he hand,
plan s a e known o cap u e ace elemen s and can be used in biomoni o ing s udies
(Ma ke , 1996; Rossini Oli a and Valdés, 2003). Se e al s udies ha e been done on he
con amina ion by pa icula e ma e in he ci y (Janssen e al., 1997; Monaci e al., 2000;
Fe nández e al., 2001), and on he use ul o plan species as biomoni o s (Ba ghigiani e al.,
1991; Sawidis e al., 1995; Aksoy and Öz ü k, 1997; Ba gagli e al., 2003; Rossini Oli a and
Valdés, 2003, 2004, Rossini Oli a & Rau io, 2004). Ne e heless, whe he a ela ion be ween
PM10 composi ion and ace elemen s in plan s exis s was insu icien ly s udied. An a emp
o s udy he ela ion be ween Hg con en in osema y and b oom lea es, Abies alba needles
and pine needles in ela ion o hose in a mosphe e was done in I aly (Ba ghigiani e al.,
1991; Ba ghigiani and Bauleo, 1992; Ba ghigiani and Ris o i, 1995).
Se ille is he mos densely popula ed ci y o Sou he n Spain, wi h 716.937
inhabi an s, si ua ed only 10 m abo e he sea le el on an ex ensi e plain c ossed by he
Guadalqui i Ri e . The ci y has a Medi e anean clima e wi h an annual a e age
empe a u e o 19 °C and ain all o 580 mm. T a ic ep esen s he mos impo an pollu ion
sou ce since indus ies a e a he sca ce and me allu gic p oduc s ep esen only 15% o his
indus ies. Impo an con ibu ions o c us al pa icles in Se ille and in he sou h o Spain a e
well known by he local scien i ic communi y and he Spanish and he egional go e nmen s.
The aim o his s udy was o compa e ace elemen le els be ween he con en in
PM10 and lea es o Ne ium oleande L. (oleande ) and Lan ana cama a L. (lan ana) o
e alua e i a di ec co espondence be ween he me al con en o pollu an s in ai and in plan
exis s. These species we e chosen because hey a e widesp ead in he ci y o Se ille (S o
Spain), bo h a e pe ennial and Oleande is know o be an use ul biomoni o (Sawidis e al.,
1991; Aksoy and Öz ü k, 1997; Rossini Oli a and Valdés, 2003). In addi ion, he
composi ion o inhalable a mosphe ic pa icles was s udied o es ablish he a mosphe ic
con amina ion le el o Se ille.
MATERIALS AND METHODS
Samples o pa icles smalle han 10 µm (PM10) and adul lea es o N. oleande and
L. cama a we e collec ed simul aneously in h ee sampling si es om au umn 2003 o sp ing
o 2004. A o al o 22 samples o PM10 we e collec ed oge he wi h he samples o bo h
species o plan s a he same day. Samples we e usually aken on di e en days o he week,
ob aining samples o e e y day o he week, he e o e, in di e en me eo ological and
en i onmen al condi ions. Sampling pe iods o ai bo ne pa icles we e om 0:00 a.m. o 23:59
p.m. Acco ding o he local au ho i ies in o ma ion, one o he sampling si e ep esen s he a ea
o Se ille a ec ed by he highes ehicula emissions (To neo, TRN) and he o he is a less
con amina ed a eas wi h low a ic densi y (Po eni , POR). The hi d sampling si e (BLK),
was chosen as a con ol si e; i is loca ed a abou 40 Km a om Se ille, in Sie a Mo ena,
away om any a ic emissions bu wi h simila clima ic cha ac e is ics.
Se e al b anches om h ee heal hy-looking plan s we e cu o om sides o he
canopy a 1.5 m o N. oleande and a 0.5 m o L. cama a. Once in he lab, abou 300 g o
esh ma u e lea es (including he pe ioles) we e mixed o p o ide a unique sample. Samples
we e d ied o e nigh a 70 °C, g ound o a ine powde .
A mosphe ic pa icula e ma e was collec ed wi h a s anda d high- olume sample MCV,
model CAV-A/HF equipped wi h a PM10 inle (MCV, model PM10/CAV). PM10 inle was
used o collec all pa icles less han 10 mic ome e s (aed) and he MCV ma k was in e -
compa ed wi h he e e ence PM10 inle o he EN 12341 Eu opean No m. The sampling
sys em was se a a low a e o 68 m-3h-1, acco ding manu ac u ing speci ica ions. Glass ib e
il e s we e used as collec ion media (Wha man GF/A, 24.5 x 20.3 cm). Be o e he analysis
o ai bo ne samples, il e s used o pa icle collec ion we e d ied, be o e and a e weighing,
in a high- olume desicca o unde 3% R.H. o a 48-h pe iod. Ope a ions o conse a ion,
d ying and weighing ook place in a da k oom a 505 % R.H, a empe a u e o 201 ºC was
con olled by a hea ing-cooling sys em. Ca e was aken in handling he sampled il e s and
plan s in o de o a oid con amina ion p oblems using a e ical lamina ai low cabine wi h
a HEPA il e INDELAB, model IDL-48V, he eby ensu ing ai cleanliness s anda ds o class
100 acco ding o he US-Fede al S anda d 209E (FED.STD-209E, 1988).
Fil e s and plan s we e placed in closed e lon essels wi h 15 ml o HNO3-HClO4
/ (3:1) and diges ed in a mic owa e o en Milles one, model E hos 900 wi h a speci ic
diges ion p og amme. A e diges ion and cooling, sample solu ions we e ans e ed in o
s e ile ubes.
Since indi iduals blanks we e no a ailable o each il e used o sampling, a se o
unexposed il e s o each ype o il e was analysed as blanks using he same p ocedu e used

o samples. The mean unexposed il e s alue we e de e mined and sub ac ed om each
sample o ob ain he bes es ima e o each elemen in he pa icula e ma e .
The accu acy o he de e mina ions we e p e iously checked by analysis o a NIST e e ence
ma e ial (SRM 1648) o ai bo ne pa icles and a GBW 07604 (popla lea es) e e ence
ma e ial o plan s. A gold solu ion was used as in e nal s anda d o spike samples, blanks
and e e ence ma e ial and o check he co ec quan i ica ion o he elemen s. P e iously, i
was s udied ha gold was no p esen ed in u ban ai and plan s and i s p esence wi hin he
e lon essel had no in luence on analy ical de e mina ions o he o he me als. All ul apu e
eagen s and s anda ds we e supplied by Me ck and dis illed wa e was o Milli-Q g ade
ob ained om a Wa e s-Millipo e appa a us, model Plus. The concen a ion o Ba, Cu, Fe,
Pb, Mn, Ni, Ti and V ha e been de e mined simul aneously by induc i ely coupled plasma
a omic emission spec ome y (ICP-AES), using a Fisons-ARL 3410 sequen ial mul i-
elemen ins umen . The s anda d ope a ional condi ion o his ins umen a e summa ised as
ollows: he ca ie gas, coolan gas, and plasma gas is a gon a 80 psi o p essu e, he ca ie
gas and plasma gas low a e a e 0.8 l min-1, he coolan gas low a e is 7.5 l min-1 and he
in eg a ion ime is 1 s. One mini- o ch consumes a gon gas a a adio equency powe o 650
W. Consequen ly, i is capable o consuming a ew millili es o sample a a low a e o 2.3
ml min –1. In o de o a oid e o s in esul s due o he e ec p oduced by he di e en
iscosi y and densi y o he acid ma ix o samples on he ICP nebulisa ion, calib a ion cu es
ha e been made wi h he same acid ma ix.
S a is ical analyses we e ca ied ou using he CSS:STATISTICA (S a So ) so wa e
package. Summa y s a is ics we e used o calcula e he mean alues and s anda d de ia ion
and, one-way analysis o a iance (ANOVA) o de ec signi ican di e ences o mean
con en be ween si es and species (p0.05). Rela ionship be ween me al concen a ion in
pa icles and in bo h species was es ed by he Pea son co ela ion coe icien s ( ). The
in e p e a ion o he co ela ion esul s consis in assigning a common sou ce o he me al
ac ions co ela ed, in ela ion o he main sou ces o his ci y, such as he a ic, he land
pa icles and he indus ial ac i i ies. In ou case, when one me al inc eases hei
concen a ion in he ai , pa allelly o o he me al h ough di e en sampling si es and
di e en days, one can conside ha hese me als come om he same emission sou ce.
Because a high co ela ion coe icien does no necessa ily imply linea i y, linea i y was
e i ied by g aphical examina ion. Any non-linea case was disca ded.
RESULTS AND DISCUSSION
The s udied elemen s mean concen a ions ound in PM10 and in plan s in di e en
si es a e epo ed in Table 1. The alues o pa icles me al concen a ion o all sampling si es
can be conside no dange ous o human heal h, since hei alues showed lowe
concen a ions han hose indica ed by he Uni Risk (UR) alues on cance o he WHO
(Wo ld Heal h O ganiza ion). Only o Ni, he mean alue was close o he UR limi (2.5 ng
m-3), indica ing he po en ial isk o ha m ul o his elemen . The esul o Cu and Pb is also
acco ding o Ross (1994) c i e ia, since he concen a ion o hese wo elemen s in bo h
s udied species did no exceed he uppe limi o alues conside ed oxic o p esen in
con amina ed plan s.
The ANOVA esul s showed ha he ai pollu an concen a ions we e no
signi ican ly di e en be ween he a ious sampling si es, excep o Mn, Cu and Fe. The Mn
concen a ion was signi ican ly low in he con ol si es espec o POR and TRN and, his
was also he case o Cu and Fe only compa ing o TRN. The TRN si e ep esen s he zone o
he ci y mo e in luenced by he ehicula a ic as i was abo e men ioned, sugges ing ha
a mosphe ic pollu ion by bo h elemen s is highe due o he a ic o soil e-suspension
(Fe nández e al., 2004).
In ela ion o he plan s, in bo h species he mean concen a ion o Pb, Ni, Mn, Cu
(excep o lan ana en POR), Fe (excep o lan ana in POR) and Ba we e below han he
e e ence alues gi en by Ma ke (1996). On he o he hand he V and, mainly he Ti mean
alues in bo h species we e always highe han he e e ence alues, including he samples
om he con ol si e, sugges ing ha hey come mainly om he soil.
Lea es o oleande showed a mean concen a ion o Cu, Pb, Fe and Ni lowe han ha
epo ed by Rossini Oli a and Valdés (2003) in Pale mo. Sawidis e al. (1995) ound highe
alues o Cu and Pb in con ol si e and in ci y oads o Tessaloniki han hose ound in Se ille
and he same esul s we e epo ed by Aksoy and Öz ü k (1997) in An alya (Tu key),
con i ming he da a o PM10 abo e discussed.
In he case o N. oleande , di e ences o Pb, Ni and V concen a ion in he con ol
si e wi h espec o he o he si es we e no ind, whils a lowe Cu, Fe and Mn concen a ion
we e ound in he con ol si e. Fo L. cama a pollu an concen a ions we e signi ican ly
di e en be ween he a ious sampling si es, excep o Ni. The Pb, Ba, Cu, Fe and V
concen a ions we e signi ican ly lowe in he con ol si es only wi h espec o POR and, Mn
concen a ion in he con ol si e was signi ican ly highe han in bo h si es; signi ican ly
highe Ba and V concen a ions in he con ol si e wi h espec o TRN we e also ound. In
addi ion, he Ti concen a ion in he con ol si e was signi ican ly highe han in he o he
si es. This means ha lan ana is no use ul as biomoni o since i does no e lec s he
pollu ion en i onmen al condi ions.
Co ela ion be ween con en in PM10 and plan s was only ound in one species, N.
oleande and only o Cu and Fe (Fig. 1), indica ing ha oleande can be a mo e use ul
biomoni o s o hese wo elemen s han L. cama a. This is in acco dance wi h Jones e al.
(1985) s a ing ha high posi i e associa ions be ween he pollu an concen a ions in plan s
and hei en i onmen sugges he po en ial use o pollu ion moni o ing in gene al and o
he examined me als in pa icula . Ba ghigiani and Ris o i (1995) did no ound any
ela ionship be ween Hg concen a ion in osema y and b oom lea es and in ai , indica ing
ha one o he easons could be he low Hg ai concen a ion. Since a mosphe ic pollu ion in
he sampling si es can be conside ed low, his can be he eason because he absence o
ela ionship in lan ana, indica ing he e o e ha i is less sensible han oleande o de ec
a mosphe ic pollu ion a ia ions by Cu and Fe.
Al hough lan ana has a oughe lea su ace han oleande , ha i is known ha his
cha ac e is ic inc eases he dus deposi ion, his s udy shows ha his p ope y look likes han
does no a ec s he capaci y o e ain and ap mo e pollu an s, since he ANOVA esul s
showed ha signi ican di e ences (p0.05) be ween he wo species only exis s o Cu and
Fe con en s, in addi ion oleande lea es accumula es mo e Ba han lan ana. On he o he hand
like Fa me epo ed (1993) ha his cha ac e is ic is impo an o pa icle g ea e han 10
m, he e o e his can be he eason because he null co ela ions ound in lan ana.
CONCLUSIONS
The a mosphe ic pollu ion by me als o he Se ille Ci y can be conside ed low and gene ally
no dange ous o human heal h excep o Ni. I on and Cu we e specially di e en ia ing
be ween he con ol si e and he u ban si es. These wo me als we e he only co ela ed
be ween pa icles and oleande lea concen a ions, indica ing ha his species is mo e use ul