iba applications to cultural heritage and environmental...
TRANSCRIPT
�������� � �
������ �� ��� ������������
��� ����������� ������������
�� � �� ����� ����
��� ������������ �� �������� �������� ��� ������������� ��������
���� ������
����������� �� �������� �����
IBA applications to Cultural Heritage and environmental problems
Pier Andrea MandòDpt. of Physics, University of Florence, Italy
and
Istituto Nazionale di Fisica Nucleare, Sezione di Firenze, Italy
Laboratorio di tecniche nucleari per i Beni Culturali
http://labec.fi.infn.it
IBA features that make them ideally suitable for C.H. and environmental studies• Very large cross sections (PIXE in particular) very low
beam currents (therefore no damage), short time needed for analysis, and great absolute sensitivity (therefore
analysis of very low target mass)• Non destructivity measurements can be repeated, also
with other techniques• External beams no need of picking up samples, large
objects, ease of handling the “targets”• Possibility of easily varying beam energy, intensity and
size, in order to find the best experimental conditions for the specific problem
• Complementary information obtained by the different IBA techniques, easily implemented in the same set-up
σ X(b
arn)
σ X(b
arn)
increasing Z
sezioni d'urto di produzione X (serie K)
0,1
1
10
100
1000
10000
1 2 3 4 5energia protoni (MeV)
Na
Ag
Zr
As
Fe
Ca
sezioni d'urto di produzione X (serie L)
0,1
1
10
100
1000
10000
1 2 3 4 5energia protoni (MeV)
Pb
Ba
Ag
Zr
Ta
X ray production cross sections (KX ray production cross sections (K--series)series)
X ray production cross sections (LX ray production cross sections (L--series)series)
Proton energy (MeV)Proton energy (MeV) Proton energy (MeV)Proton energy (MeV)
barns barns
IBA + AMS accelerator facilityin Florence
Tandetron Accelerator3 MV terminal voltage
IBA dual
source injector High-energy AMS
spectrometer
Multi-sample AMS injector
IBAbeamlines
multipurpose IBA vacuum chamber
electrostatic chopper
external beam for archaeometry
external beam for aerosol analysis
external µ-beam
multi-angle scattering chamber
PIXE-induced XRF
operationalunder installationplanned
Overall view of the Florence Tandetron accelerator hall
The beamlines so far installed for IBA
Gamma-ray detector(PIGE)
X-ray detectors(PIXE)
External-beam set-up for the analysis of paintings and other art objects
An essential facility to perform IBA An essential facility to perform IBA especially in the field of Cultural Heritage especially in the field of Cultural Heritage
1 cm
the external beam setthe external beam set--upup
With an external beam you can investigate in a non-destructive way the
complete quantitative composition of any material you may be interested in
...glazed terracottas,
External PIXE analysis of the “Ritratto di fanciullo”by Luca Della Robbia –before restoration at the Opificio delle Pietre Dure
in Florence
Analysis of ancient glass,
Glass mosaic tesserae found in excavations at
Villa Adriana, Tivoli
1 cm
External PIXE-PIGE analysis of the glass tesserae from Villa
Adriana
External-beam PIXE analysis of the frontispiece of Pl.16,22 (XV century, Biblioteca Laurenziana in Florence)
...ancient illuminated manuscripts,
...historical documents,
Inks in Galileo’s manuscripts (Florence National Library) analysed by external PIXE
...drawings,
PIXE-PIGE analysis of a drawing on prepared paper, by Leonardo or his school
...paintings on wood or canvas,
Differential PIXE and PIGE analysis of the “Ritratto Trivulzio” by Antonello da Messina
…or analysis of aerosols collected on filters
Particulate Matter (PM) from streaker samplers (1-hour
resolution)
PM from daily samplers
Typical experimental conditions in applications to C.H.
• proton beams, 1 ÷ 5 MeV• 5÷50 pA currents, 100÷200 s runs• 0.1÷1 mm beam size• two X ray detectors
one for lower-Z elements, covering a small solid angle; He flow for minimising absorption the other for higher-Z elements, covering the largest possible solid angle and with proper absorbers to cut the high rate of low-energy X rays
• a gamma ray detector
Two-detector PIXE setup,collimated
external beam
10
100
1000
10000
100000
0 5 10 15 20 25 30 35X ray energy (keV)
coun
ts/(
g/cm
2 )(C
)
K X-rays BIG
L X-rays BIG
K X-rays SMALL
L X-rays SMALL
X ray detection efficiencies in a two-detector setup
Analysis of documents of historical interest
A letter of Galileo during PIXE A letter of Galileo during PIXE analysis with the external beam at analysis with the external beam at the Florence acceleratorthe Florence accelerator
(INFN FI, Bibl.Naz. FI, MPI Berlin)
PIXE measurements to quantitatively determine ancient inks composition
Important contribution to the chronological reconstruction of
Galileo’s hand-written notes about motion
Comparison of ink composition in the notes (which are not
dated) with that in dated documents (letters, etc.)
Some folios from Ms.Gal.72
(Bibl. Naz. Firenze)
A precious “database” of dated inks: records of money transactions in Ms.Gal.26
0
2 0 0
4 0 0
6 0 0
8 0 0
1 0 0 0
1 2 0 0
1 4 0 0
Cont
eggi
F e
F eC u Z n
P bP b
P bM n
M s.G a l.7 2 f .1 2 8
0
5 0 0
1 0 0 0
1 5 0 0
2 0 0 0
2 5 0 0
Con
tegg
i
F e
F e Z nC uM n Z n P b P b
M s.G al.2 6 f .2 9 v
0
5 0 0
1 0 0 0
1 5 0 0
2 0 0 0
2 5 0 0
3 0 0 0
4 0 0 0 5 0 0 0 6 0 0 0 7 0 0 0 8 0 0 0 9 0 0 0 1 0 0 0 0 1 1 0 0 0 1 2 0 0 0 1 3 0 0 0 1 4 0 0 0
E n e rg ia (e V )
Con
tegg
i
M n
F e
F eC u Z n Z n P b
M s.G a l.1 4 f .2 7 r
P b
Discriminating between different Discriminating between different inks withinks with PIXEPIXE
v(s), v(t), s(t)
Folio 128the “starting point”
...che il grave cadente naturalmente vada continuamenteaccrescendo la propria velocità....
...secondo che accresce la distanza dal termine onde si partì....
““DatingDating”” f.128f.128
v(s), v(t), s(t)Folio 164v - the “final result”
...sunt inter se ut radices distantiarum...
Folio 164vFolio 164v
First part
two final lines
Folio 164 v Folio 164 v –– comparison between the comparison between the two propositionstwo propositions
0.00
5
0.01
5
0.02
5
0.03
5
0.04
5
0.05
5
0.06
5
4^ p
rop.
3^ p
rop.
02468
101214161820
num
ero
di c
asi
rapporto in peso
164v Cu/Fe
0.2
0.4
0.6
0.8
1
1.2
4 ̂p
rop.
3̂ p
rop.
02468
1012141618
num
ero
di c
asi
rapporto in peso
164v Cu/Zn
0.01
0.03
0.05
0.07
0.09
0.11
4^ p
rop.
3^ p
rop.
0123456789
10
num
ero
di c
asi
rapporto in peso
164v Pb/Fe
Connection between different folios (91v, 152r)
Connection between f.91v and f.152r
0,00
5
0,01
0,01
5
0,02
0,02
5
0,03
0,03
5
0,04
0,04
5
0,05
0,05
5
0,06
0,06
5
0,07
0,07
5
0,08
152r
91v
corr
91v
text
0
0,5
1
1,5
2
2,5
3
3,5
4
4,5
5
nr of cases
weight ratio
91v-152r Cu/Fe
0,00
25
0,00
5
0,00
75
0,01
0,01
25
0,01
5
0,01
75
0,02
0,02
25
0,02
5
0,02
75
0,03
0,03
25
0,03
5
0,03
75
0,04
152r
91v
corr
91v
text
0
1
2
3
4
5
6
7
nr. of cases
weight ratio
91v - 152r Pb/Fe
0,1 0,2 0,3 0,4 0,5 0,6 0,7 0,8 0,9 1 1,1 1,2 1,3 1,4
152r
91v
corr
91v
text
0
1
2
3
4
nr. of cases
weight ratio
91v - 152r Zn/Cu
f. 179r
f. 179r
0,00
5
0,01
5
0,02
5
0,03
5
0,04
5
0,05
5
0,06
5
0,07
5
text
+dra
w.
und.
+glo
ssae
0
2
4
6
8
10
cases
weight ratio
Cu/Fe
0
1
2
3
4
5
6
7
8
cases
weight ratio
P b/Fe
0
2
4
6
8
10
12
cas es
weig ht ra t io
(Cu/Zn)/10
f. 179v
f. 179v
0,00
5
0,01
5
0,02
5
0,03
5
text
vto
p a1 a1 a2
0
1
2
3
4
5
cases
weight ratio
Cu/Fe
0
1
2
3
4
5
cases
weight ratio
Pb/Fe
0,1
0,3
0,5
0,7
0,9
1,1
text
v
a1
01234567
cases
weight ratio
Cu/Zn
Simultaneous PIXE and PIGE analysis• In most cases, great advantage in combining
PIXE and PIGE:– by γ rays, light elements are easily detected,
while their detection through X rays is impossible at all or problematic
External beam set-up equipped with:
two Si(Li) dets for X rays
Ge(Hp) det for γ rays
Glass mosaic tesserae from wall decorations in Villa Adriana (Tivoli)
Surface inhomogeneities due to
glass alterations
1 cm
Sodium detection and quantification is of the greatest importance to characterise ancient glass
In Europe, two typologies are found, depending on the component used to lower the melting temperature of silica:• natron (sodium carbonate) glass with high Na2O and low K2O
content (as in Roman and Early Middle Age glass)• Ashes from plants glass with high K2O content (later periods)
PIXE and PIGE analysis of ancient glass• Surface alterations and crusts prevent from detecting
sodium with PIXEColoured but more opaque zone(apparently with no alteration however)
“Freshly” cut zone
0
100
200
300
400
0.8 1 1.2 1.4 1.6 1.8 2Energy (keV)
Nor
m. C
ount
s (a
.u.)
Na
Al
Si
0
200
400
600
800
1000
1200
1400
0 1 2 3 4 5Energy (keV)
Nor
m. C
ount
s (a
.u.)
CaAl
Si
Na
0
100
200
300
400
0.8 1 1.2 1.4 1.6 1.8 2Energy (keV)
Nor
m. C
ount
s (a
.u.)
Na?1.04 keV
AlSi
0
200
400
600
800
1000
1200
1400
0 1 2 3 4 5Energy (keV)
Nor
m. C
ount
s (a
.u.)
CaAl
Si
Na?
Na doesn’t seem to be present...
Instead, using PIGE (Na characteristic γ-rays at 441 keV)
Na can be quantified by PIGE even in the presence of surface alterations, with no need of picking up samples!
More opaque zone
0
20
40
60
80
100
120
200 300 400 500 600Energy (keV)
Nor
m. C
ount
s (a
.u.) Na 440 keV
Freshly cut zone
0
20
40
60
80
100
120
200 300 400 500 600Energy (keV)
Nor
m. C
ount
s (a
.u.) Na 440 keV
High - Na2O were found (from 10% to 20%, depending on colour), with a composition compatible with that in typical Na-Ca Roman glass
PIXE and PIGE analysis of ancient glass
Analysis of paintings on wood on canvas
Understanding the “secrets”of painting
techniques of famous artists and/or reconstructing the
history of a specific painting (possibility to be a fake,
previous restorations, etc.)
PIXE – PIGE analysis of“Ritratto Trivulzio” byAntonello da Messina,
at LABEC, Florence
The protective varnish layer on
paintingstwo problems
1) discriminating components in the varnish
from those in paint and substrate layers
2) PIXE detection of light elements in the underlying layers (X ray absorption)
Stratigraphic analysisIf the target is not homogeneous in depth
traditional PIXE does not provide information about the stratigraphic layout of elements.
Indeed, when the beam penetrates through different layers, their contributions are added up in the spectrum
with no possibility to discriminate where X rays originate from
substratepreparation
paint layer(s)varnish
incident beam
Differential PIXE
substrate
preparationpaint layer
varnishConsists in performing measurements on
the same areawith beams of different energies
By comparing X ray spectra taken at different energies, stratigraphic information can be obtained
E1
E1
E2
E2 <
E3
E3 <
At different energies, beam ranges are different
probed depth also changes
Differential PIXE
The number of layers and their thickness are not known a priori, therefore the most suitable choice of beam energies
to discriminate layers is not obvious
the same element may be present in different layers
X ray production cross sections change significantly, and in different ways for the different elements, at varying beam energies
E1E2E3 E3 < E2 < E1
Quantitative analysis more difficult
Analysis is made complex by several factors:
However, diff. PIXE is often a “unique” tool to learn about elemental depth distribution in paintings without picking up samples
PIXE spectra at different energies
Prepared sample:wood substrate, chalk (CaSO4) preparation,
lapislazuli paint 0
300
600
900
1200
0 1 2 3 4 5E (keV)
Cou
nts
No degrader
Al
CaK
Si
S
Na
0
300
600
900
1200
0 1 2 3 4 5E (keV)
Cou
nts
75 µm Upilex
3 MeV
1.7 MeV
Na
AlSi
S
KCa, S
Na, Al, Si, S, K
Leonardo Madonna dei fusi
ex-Reford version
(private collection)
Oil painting on wood, 50 x 36
Presumably painted in 1501
Varnish compositionUsing differential PIXE, the varnish composition was evaluated from the spectra collected at the lowest beam
energy, when protons do not reach the underlying paint and preparation layers
0.0
0.1
0.2
0.3
0.4
0.5
0.6
2000 4000 6000 8000 10000 12000 14000 16000Energy (eV)
Cou
nts/
nC
Ca
Fe
Cu
Zn
b)
Varnish spectra for a) lower-Z elements and b) higher-Z elements
From the comparison of differential PIXE spectra, it was also possible to estimate the thickness of the varnish
layer: from ~30 to ~ 50 micron
0.00.20.40.60.81.01.21.41.61.8
0 1000 2000 3000 4000 5000 6000 7000 8000Energy (eV)
Cou
nts/
nC
ClSiAl
Ca
S
Fe
Na
Mg
P
K
a)Ar (from air)
Element ConcentrationNa, Cl, Ca ~ 1‰ Fe ~ 0.5‰ Al, Si, S 0.5-1‰ Mg, P, K 0.2-0.5‰ Ti, Cu, Ba ~ 0.1‰ Zn 0.1-1‰
0
5
10
15
20
25
30
2000 4000 6000 8000 10000 12000 14000 16000Energy (eV)
Cou
nts/
nC
2.8 MeV
Fe hematite?
Hg use of cinnabar as red pigment
Pb lead-white (in the paint layer? in the preparation substrate? in both?)
Ca
Fe Hg
Pb
a a
b
c
d
0.0
0.5
1.0
1.5
2.0
2.5
2000 4000 6000 8000 10000 12000 14000 16000Energy (eV)
Cou
nts/
nC
2.3 MeV
0.0
0.2
0.4
0.6
0.8
1.0
2000 4000 6000 8000 10000 12000 14000 16000Energy (eV)
Cou
nts/
nC
2 MeV
0.0
0.1
0.2
0.3
0.4
0.5
0.6
2000 4000 6000 8000 10000 12000 14000 16000
Energy (eV)
Cou
nts/
nC
1.8 MeV
b
d
c
HgPb
Hg
Pb
Ca
Fe
Incarnato
Ca and Fe peaks are entirely accounted for by their abundance in the varnish.
An estimate of the paint layer thickness is obtained: only 15 – 20 µm!!
The protective The protective varnish layer on varnish layer on
paintingspaintingstwo problemstwo problems
1) discriminating 1) discriminating components in the varnish components in the varnish
from those in paint and from those in paint and substrate layerssubstrate layers
2) detecting light elements 2) detecting light elements in the underlying layers in the underlying layers
(X ray absorption)(X ray absorption)
simultaneous use of simultaneous use of PIGE to detect light PIGE to detect light
elementselements
0
20
40
60
80
100
2000 4000 6000 8000 10000 12000 14000 16000Energy (eV)
Cou
nts/
nC
Mountains, pale blue, original
Pb in the PIXE spectrum mainly derives from lead white mixed in the blue paint
PIXE spectrum PIGE spectrum
CaFe
Pb
0
1
2
3
4
340 360 380 400 420 440 460 480 500 520Energy (keV)
Cou
nts/
nC
Gown of the Virgin, dark blue, restoredCo (cobalt blue) and Zn (zinc white) identify this as a restored detail. The small Pb peaks are due to the beam halo hitting original blue areas around (as seen by a lateral scan)
0
2
4
6
8
340 360 380 400 420 440 460 480 500 520Energy (keV)
Cou
nts/
nC
0
5
10
15
20
25
30
2000 4000 6000 8000 10000 12000 14000 16000Energy (eV)
Cou
nts/
nC
2.8 MeVZn
FePb
Co
Ca
440 keV(Na)
Identification of lapislazuli by PIGE
One can also produce an externalmicrobeam
Through collimation, well defined beams of no less than 100-200 µm can be produced
Smaller-size beams are obtained with strong focusing using lenses (quadrupole multiplets)
External microbeam set-up
Si3N4 exit window, 100 nm thicknessTarget at ~2 mm from exit windowX and γ detection systems as in standard external set-upBeam magnetic scan over the sampleMechanical scan, i.e. sample micrometric displacement in front of the beamLIST-MODE (E,x,y) acquisition element mapsMinimum beam size on target: 10 µm
scanning-IBA for Cultural Heritage: why?Details of small size or inhomogeneous structure (even
~100iµm) not always easily recognised by visual inspection
Risk of misleading information from single-spot measurements
Dramatic improvement in significance, reliability and completeness of information, using methodologies providing
“compositional maps”Scan of relatively large areas (~ some mm2) with beams around
100-200 µm, acquiring “pixel by pixel” info
risk of mixing info referring to
different materials
too broad beam
risk of analysing anomalous, non-representative
“points”
too small beam
Sub milli-beam
scanning IBA applications
to Cultural Heritage
glass surface alterations
Glass tesserae from Villa Adriana (I)
• Using the proton “sub milli”probe external set-up– ~ 80-100 micron beam size– Magnetic beam scan on samples
Coloured but more opaque zone
“freshly cut” zone
“crust”
Na γ-ray mapNa X-ray map
2 mm
Problem of detecting Na by PIXE, because of surface alterations
X ray maps from other elements…
Cu Only absorbed in the crust (you may get an idea of its thickness...)
Fe
Larger contribution from the crust
Coloured but more opaque zone
“freshly cut” zone
“crust”
Si
Absorbed,though less than Na
2 mm
Glass tesserae from Villa Adriana (II)
investigation of metal point drawings on coloured papers
Sub milli-beam
scanning IBA applications
to Cultural Heritage
VARIOUS KINDS OF METAL STYLUS TO DRAW ON PAPER( Köln, diocesan museum)
Silver stylus used by Hans Cranach(Hannover, Landesmuseum)
Rogier Van der WeydenSt. Lucas portraying the Virgin (detail)Boston, Museum of Fine Arts
PAOLO UCCELLO – STUDY OF A KNIGHT
Uffizi, Gabinetto Disegni e Stampe Metal point, lead white + earth-green
prepared paper
PISANELLOPROFILE OF A WOMAN
PARIS, LOUVREmetal point on prepared white
paper
LEONARDO DA VINCISTUDY OF A DRAPERY
ROMA, ISTITUTO NAZIONALEPER LA GRAFICA
metal point + lead whitered prepared paper
Metal-point drawings on prepared paper
Knowledge of materials is needed for conservation purposes: one is dealing with very fragile and precious works, so far little studied, and mainly from the art-historical point of view
Need of a non-destructive imagingtechnique, with a space resolution of
100-200 µm at least2 mm
Problem:
non uniform track left by the metal stylus make material identification difficult, especially when the paper is prepared using compounds of the same metal
Paper prepared with cinnabar + Pb whiteCu stylus
Cu map
2 mm
Pb map
Pb stylus
2 mm
investigation of iron gall inksto discriminate their different compositions
Sub milli-beam
scanning IBA applications
to Cultural Heritage
Iron gall inks1 cm
XVII century document from
State Archive in Florence
1 mm1)
Cu X map
S X map
Fe X map
Min.
Max.
Possibility to select “good”areas, from which extracting quantitative reliable information
b
a
Fe
a
b
Si
a
b
Ca
0
200
400
600
800
1000
1200
1400
2 3 4 5 6 7 8 9 10 11 12Energy (keV)
Cou
nts
Ca
Fe
MnZn
Cu
0
100
200
300
400
500
0 1 2 3 4 5 6 7 8Energy (keV)
Cou
nts
S
K
Ca
Fe
AlSi
Na
Mg
0
50
100
150
200
250
300
2 3 4 5 6 7 8 9 10 11 12Energy (keV)
Cou
nts
Ca
Fe
Mn
ZnCu
0
100
200
300
400
500
0 1 2 3 4 5 6 7 8Energy (keV)
Cou
nts
S
Ca
Fe
AlSi
Mg
a)
a)PIXE
b)
b)
0
10
20
30
40
50
60
70
200 300 400 500 600 700 800Energy (keV)
Cou
nts
0
10
20
30
40
50
60
70
200 300 400 500 600 700 800Energy (keV)
Cou
nts
Na441 keV
PIGEa) b)
off-line spectra reconstruction
2 mm
IBA application to air pollution monitoring
• This field has historically been among the first where PIXE found useful application
• Still now, largely used by PIXE labs in the world
• The analytical problem is the scarce quantity of target material (aerosol deposited on filters), and the large PIXE cross sections help to achieve very
good detection limits
I
• Beam currents used are much higher than for C.H. (up to 20÷30 nA depending on the problem)
• Beam size is normally larger (up to some mm2)
• PIGE is also standard for light elements detection
• Using non external set-ups, PESA (Particle Elastic Scattering Analysis) also provides very useful
information
IBA application to air pollution monitoring
II
Why studying aerosols
• effects on environment and health• origin (sources)
size andcomposition
• to better understand the great physical processes in the atmosphere (study of climate changes etc.)
• to evaluate pollution levels and identify pollution sources in urban and industrial areas
It is important to measure concentration and composition of the different size fractions, and their time
behaviour
Aerosol (also referred to as PM ≡ Particulate Matter) is continuously monitored by local authorities in the main urban areas, but in most cases only average daily concentration of PM10 (PM with size below 10 µm) is measured: compositional analyses and size-fractionated samplings are not
routinely performed
Composition must be obtained through techniques providingfast runs (many samples!)
high sensitivity (conc. << mg/cm2)multi-elemental analysisnon destructive analysis
Data analysisEvaluation of air quality, correlation to other pollutants and
meteorological parameters, comparison among sites, Identification of pollution sources and of their relative weight
Sampling• time resolution matching specific demands
• size fractionation
thinthin depositsdeposits ((∼∼1010--300 300 µµg/cmg/cm22)) ofof aerosol aerosol onon filtersfilters or impactorsor impactors
LargeLarge quantity of data are collectedquantity of data are collected (concentration of many (concentration of many elements/compoundselements/compounds in in air, for a large number of samples)air, for a large number of samples)
SamplingPumping volumes of air:
Low Volume (< 0.05 m3 /min)High Volume (∼ 1 m3/min)
Continuous o cumulative samplers
Size fractionation:single mode (all aerosol below a certain size is collected)
multi-mode (often just bimodal to separate fine and corase fraction)
PM is collected:by impaction (inertial samplers)by filtration through membranes
A multi-stage inertial sampler providing size-fractionation
Samplers
Ingressoaria
Tubo di raccordo
Filtro
Pompa
Testa di prelievo
• Sampling heads for PM1, PM2.5 or PM10• One-day resolution• Mass of deposit obtained either by weighting or
by β attenuation
47 mm
A continuous sampler (streaker)
providing bimodal size-fractionation
100 mm
PM2.5
PM2.5÷10
MOTOR
NUCLEPORE FILTER
AIR OUTLET
KAPTON FILM
PRE-IMPACTOR
AIR INLET
MOTOR
NUCLEPORE FILTER
AIR OUTLET
KAPTON FILM
PRE-IMPACTOR
AIR INLET
Motore
Impattore(kapton)
Filtro (nuclepore)
Pre-impattore
Ingresso aria
Pompa
The streaker samplerpumping orifice
Fine-fraction (PM2.5) filter from a streaker
PM10, ∼280 µg/cm2 PM10, ∼30 µg/cm2
2 2 µµmm 2 2 µµmm
10 10 µµmm 10 10 µµmm
particulate substrate
Particle beam
• Particle Induced X-ray Emission (PIXE)• Particle Induced Gamma-ray Emission (PIGE)• Particle Elastic Scattering Analysis (PESA):
FS and BS (Forward and Back Scattering)
detector
characteristic radiation
IBA on aerosols
Back-scattered protons (BSP) from target
131 eV148 eVRESOLUTION(2.3 keV)
6%14%PILE-UP
0.1738BSP (cts/nC)
WITHDEFLECTOR
WITHOUTDEFLECTOR clearance
permanent magnets
Nd-Fe-B (0.5 T)
proton deflector
• non destructive
only elements withZ>10 are detected
X ray absorption (< 1keV)
PIXE for aerosol composition
• multi-elemental
• high absolute sensitivity (MDL 1-10 ng/m3 in 5-10 min runs)
• quantitative
• no sample pre-treatments needed
• scanning on samples from streakers
• single-particle analysis is possible
aerosol sample analysis is the main activity in about ¼ of PIXE laboratories in the world
Typical X ray spectra from aerosols
0
1000
2000
3000
4000
5000
6000
7000
2 4 6 8 10 12 14
cont
eggi
/can
ale
Ca
Ca
Fe
FeCu
ZnCu
Zn
Pb+As
As
Pb
Ni
MnTiCrV
Br
Br Pb+Sr
ZrK
x 30
0
200
400
600
800
1000
1200
1400
1600
0 1 2 3 4
Na
MgAl
Si S
Cl
K
Ca
Ca
lower-Z elements higher-Z elements
PIXE-PIGE set-up for ‘one-day’ samples
FilterFilter--holder wheelholder wheelVacuum beamlineVacuum beamline
Si Si ““SMALLSMALL””(X < 6 keV)(X < 6 keV)
Si Si ““BIGBIG”” (X > 5 keV)(X > 5 keV)
Ge (Ge (γγ))
BackscatteredBackscatteredproton proton
deflectordeflector
He flow He flow pipepipe
Faraday Faraday cupcup
X
FC
Beam spot covers a filter sector corresponding to 1 hour sampling
p
• The deposited aerosol composition is determined “point by point” by PIXE and PIGE
• One can thus deduce aerosol composition in atmosphere “hour by hour” during sampling period
• Typically, sensitivity for the detectable elements comes out to be of the order of ng/ m3
Set-up for ‘one-hour’ samples
PIGEHPGeHPGe detectordetector
Energia (keV)
Num
ero
di c
onte
ggi
0
100
200
300
400
500
600
0 50 100 150 200 250 300 350 400 450 500
110( F)
197( F)
441( Na)23
19
19
Thin target approximation
• Negligible self-absorption(γ rays)
• Generally, non-negligible croos section changes
REAZIONE E γ (keV) 19F (p, p’γ) 19F 585, 1369
23Na (p, p’γ) 23Na 441 25Mg (p, p’γ) 25Mg 24Mg (p, p’γ) 24Mg
585
1369 27Al (p, p’γ) 27Al 844, 1014 28Si (p, p’γ) 28Si 1779 31P (p, p’γ) 31P 1266
110, 197
0
25
50
75
100
2650 2750 2850 2950 3050 3150 3250
Na - 441 keV
0
5
10
15
20
2650 2750 2850 2950 3050 3150 3250
Al - 1014 keVAl - 844 keV
0
3
6
9
12 Si - 1779 keV
2650 2750 2850 2950 3050 3150 3250
0
4
8
12
16
2650 2750 2850 2950 3050 3150 3250
P - 1266 keV
0.0
1.5
3.0
4.5
6.0
2650 2750 2850 2950 3050 3150 3250
Mg - 585 keVMg - 1369 keV
2650 2750 2850 2950 3050 3150 3250
Proton energy (keV)
Proton energy (keV)
0.70
0.75
0.80
0.85
0.90
0.95
1.00
0 30 60 90 120 150 180
angolo di scattering
Fatto
re c
inem
atic
o K
12C
16O14N
56Fe
48Ti
m = 1
PESA (Particle Elastic Scattering Analysis)
larger ∆E for smaller M (light nuclei)
Larger ∆E for larger ϑ(backscattering [BS])
222
1)/(cossin)/(
)/,(⎥⎥⎦
⎤
⎢⎢⎣
⎡
++−
=≡mM
mMEE
mMKi
f ϑϑϑ
Ef
M
Mm
m
Ei
ϑ
detectordetector
0.70
0.75
0.80
0.85
0.90
0.95
1.00
0 30 60 90 120 150 180
angolo di scattering
Fatto
re c
inem
atic
o K
12C
16O14N
56Fe
48Ti
m = 1
PESA (Particle Elastic Scattering Analysis)
larger ∆E for smaller M (light nuclei)
Larger ∆E for larger ϑ(backscattering [BS])
222
1)/(cossin)/(
)/,(⎥⎥⎦
⎤
⎢⎢⎣
⎡
++−
=≡mM
mMEE
mMKi
f ϑϑϑ
Ef
M
Mm
m
Ei
ϑ
M ≤ m ⇒ only forwardscattering [FS]
0.00
0.20
0.40
0.60
0.80
1.00
0 30 60 90 120 150 180
angolo di scattering
Fatto
re c
inem
atic
o K
H
NC
O
m = 1detectordetector
0
100
200
300
1700 1900 2100 2300 2500 2700energia (keV)
cont
eggi
/ µC
0
200
400
600
800
1000
1800 2200 2600 3000energia (keV)
cont
eggi
/C
bianco
forward (ϑ = 30°)
C F
C + Fcon particolato
Η
CO
N
Α > 1
Teflon membranes
F (anelastic)
backward (ϑ = 150°)
Set-up for PESA
samplesampleSBD
(ϑ = 150º)SBD
(ϑ = 30º)
GeGe detectordetector
PESA on aerosol samples• Complement to PIXE
can detect light elements down to Hydrogen
• multielemental
• non-destructive
• no sample pre-treatments needed
• much less sensitive (sufficiently however!)
• requires set-ups in-vacuo
• difficult to find suitable filtering substrates
• more difficult quantitative analysis
Errors and MDL of elements detected by PESA
0.4~ 15 Oxygen
0.5~ 20 Nitrogen
1~ 10 Carbon
0.1~ 10 Hydrogen
MDL(µg/m3)
Error%
Element
Some examples of results
Composition of aerosol collected in the urban area of Florence, Italy
(in collaboration with ARPAT)
V.le GramsciJan 97 – Jan 98 (249 f.)July 01 – Jun 02 (276 f.)
Sesto FiorentinoDec 97 – May 98 (144 f.)
Boboli gardenGen 97 – Feb 98 (264 f.)
Hourly samplingGen – Feb 02
Hourly samplingGen – Feb 02
PM 10
0
25
50
75
100
125
01-J
ul-0
1
01-A
ug-0
1
01-S
ep-0
1
01-O
ct-0
1
01-N
ov-0
1
01-D
ec-0
1
01-J
an-0
2
01-F
eb-0
2
01-M
ar-0
2
01-A
pr-0
2
01-M
ay-0
2
01-J
un-0
2
01-J
ul-0
2
Con
cent
razi
one
(µg/
m3 )
Viale Gramsci - total PM10 concentration (2001-02)
average:average: 43 43 µµg/mg/m33
number of days exceedingnumber of days exceeding 50 50 µµg/mg/m33 : 76: 76
Al and Si concentrationAl
0.0
0.4
0.8
1.2
01-J
ul-0
1
01-A
ug-0
1
01-S
ep-0
1
01-O
ct-0
1
01-N
ov-0
1
01-D
ec-0
1
01-J
an-0
2
01-F
eb-0
2
01-M
ar-0
2
01-A
pr-0
2
01-M
ay-0
2
01-J
un-0
2
01-J
ul-0
2
Con
cent
razi
one (
µg/
m3 )
Si
0
1
2
3
01-J
ul-0
1
01-A
ug-0
1
01-S
ep-0
1
01-O
ct-0
1
01-N
ov-0
1
01-D
ec-0
1
01-J
an-0
2
01-F
eb-0
2
01-M
ar-0
2
01-A
pr-0
2
01-M
ay-0
2
01-J
un-0
2
01-J
ul-0
2
Con
cent
razi
one (
µg/
m3 )
13 Feb 97 12:00
06 Jan 98 06:00
12 Oct 97 06:00
23 Jun 97 18:00
Na and Cl concentration
Pb and Br, before leaded gasoline ban
Lunedì 26/1/98
0.00
0.05
0.10
0.15
0.20
0.25
0.30
0.00
2.00
4.00
6.00
8.00
10.0
0
12.0
0
14.0
0
16.0
0
18.0
0
20.0
0
22.0
0
Con
cent
razi
one
( µg/
m3 )
0
500
1000
1500
2000
2500
3000
3500
numero m
acchine all'ora
Hour by hour concentration of Pb during one day
Bromine and Lead from leaded gasoline
0,00
0,01
0,02
0,03
0,04
0,05
0,00
0,05
0,10
0,15
0,20
01-lu
g-01
01-a
go-0
1
01-s
et-0
1
01-o
tt-0
1
01-n
ov-0
1
01-d
ic-0
1
01-g
en-0
2
01-f
eb-0
2
01-m
ar-0
2
01-a
pr-0
2
01-m
ag-0
2
01-g
iu-0
2
01-lu
g-02
µg/m
3Br
Pb
0,0250,23Pb
0,0060,06Br
AprilJune ‘02
AprilJune ‘97
µg/m3
Pb limit by lawPb limit by law 0.5 0.5 µµg/mg/m33 yearly averageyearly average
01/01/2002:
In Italy, sale of leaded gasoline is forbidden
values detected in Florence
Comparison 1997-98 with 2001-02
0.001
0.010
0.100
1.000
10.000
Na
Mg
Al
Si P S C
lK
C
aTi V
C
rM
nFe C
uZn B
rSr Zr Pb PM
10/10
1997-19982001-2002
µgC
once
ntra
tion(µ
g/m
3 )
Comparison with 1988-89
0.310.09 2.0Jan-98
2.20.4612.6Jan-89
0.270.06 3.5Sept-97
1.130.304.2Sept-88
PbBrSPeriod
values in mg/m3
Fireworks on Dec 31
Firenze 1997-1998
K
0
2
4
6
8
10
12
boboli
gramsci
sesto
Sr
0
0.05
0.1
0.15
0.2
22-Dec-96
6-Jan-9721-Jan-975-Feb-9720-Feb-977-M
ar-9722-M
ar-976-Apr-9721-Apr-976-M
ay-9721-M
ay-975-Jun-9720-Jun-975-Jul-9720-Jul-974-Aug-9719-Aug-973-Sep-9718-Sep-973-O
ct-9718-O
ct-972-N
ov-9717-N
ov-972-D
ec-9717-D
ec-971-Jan-9816-Jan-9831-Jan-98
µ
boboli
gramsci
sesto
Con
centr
atio
n(µ
g/m
3)
Campaigns in an industrial district 20 Km west of Florence:
Montelupo Fiorentino
(ceramics, glass)
012345
3.00
15.0
03.
0015
.00
3.00
15.0
03.
0015
.00
3.00
15.0
03.
0015
.00
3.00
15.0
0
AsNaK
Montelupo Fiorentino: identification of industrial releases in connection with artistic glass production
Episodes of correlated Na, K, As peaks during the night (March ‘96)
Fine fraction
Con
centr
atio
n(µ
g/m
3)
Absolute Principal Component Analysis
Groups detected elements according to similarities of their timebehaviour
n inter-correlated variables
(concentrations)
p < n independent
variables(‘components’or ‘factors’)
explain a large part of the data variance
represent thesources of partculate
Aerosol particles maintainthe fingerprint of their source even after long-range transportation
Aerosol composition and time behaviour provide information on pollution sources
Factor loadings
suolo sale ind. 2 comb. ind. 1Na 0,16 0,92 0,09Mg 0,86 0,43 0,12 0,06 0,12Al 0,94 0,26 0,07 0,12 0,04Si 0,95 0,21 0,14 0,10 0,08S 0,14 0,94Cl 0,29 0,90K 0,73 0,23 0,31 0,13 0,39Ca 0,90 0,09 0,23 0,06 0,18Ti 0,92 0,10 0,03 0,03 0,23V 0,38 0,64 0,48Cr 0,51 0,40 0,04 0,59Mn 0,87 0,24 0,05 0,34Fe 0,89 0,06 0,17 0,03 0,36Ni 0,40 0,23 0,23 0,77Cu 0,38 0,45 0,67Zn 0,23 0,75 0,09 0,33Br 0,55 0,22 0,68Pb 0,19 0,22 0,85 0,04 0,14
The case of Montelupo Fiorentino (industrial area)
altro7% suolo
21%
sale4%
ind. 216%
comb.11%
ind. 141%
Relative contribution of the various sources to total PM10 mass (annual average, 1998)
The case of Montelupo Fiorentino (industrial area)
Relative contribution of the various sources to total PM10 mass (1997-1998)
heavy traffic area
(50 ± 2)%
(21 ± 1)%
(21 ± 1)%
(8 ± 3)%
urban park
(20 ± 2)%
(19 ± 3)% (43 ± 2)%
(18 ± 2)%
residential area
(38 ± 4)%
(32 ± 3)%
(26 ± 3)%(5 ± 4)%
TRAFFICO
COMBUSTIONE
SUOLO
ALTRO
The case of Florence (urban area)