Presentazione standard di PowerPoint

THE STARTRACK EXPERIMENTnanodosimetric STructure of hAdRon TRACKs

2nd Annual WorkshopMilano, October 2013

Valeria ConteA really complicated experiment with interesting results 1CONTENTSGENERALITIES OF TRACK STRUCTURE

THE EXPERIMENTAL SET UPThe rational of the experimentThe measuring procedure

DATA ACQUISITION AND DATA ANALYSISHow to handle experimental data

RESULTSSome interesting features of the track structure of light ions

22

1 mmEffectiveness of radiation strongly depends on the track structureTrack structure = detailed description of the interaction pattern of single particles(i) where the interaction takes place (spatial distribution of ti)(ii) which kind of interaction (local energy deposits i)(iii) what species are possibly formed

The hypothesis: The damage to segments of the DNA is initiated to a great part by ionizing processes

not yet possible experimentallyTHE TRACK STRUCTURE OF IONIZING PARTICLES33

1 mmTHE TRACK STRUCTURE OF IONIZING PARTICLESTrack structure = detailed description of the interaction pattern of single particles(i) where the interaction takes place (spatial distribution of ti)(ii) which kind of interaction (local energy deposits i)(iii) what species are possibly formed

Characterization of Particle Track Structure by Measuring the frequency distribution of the number of ionizations in nanometre-sized target volumes

not yet possible experimentallyEffectiveness of radiation strongly depends on the track structure44

The track component due to primary-particle interactions:Track-core regionThe track component due to secondary-particle interactions:Penumbra region

the track structure of energetic ions5A detector to measure the track structure characteristics in the core and in the penumbra5

20 nmThe physics of the interactionat the mm level

at the nm level

A picture of the track structure with a pixel of 20 nm1 mm61 mm6STARTRACK: THE EXPERIMENTAL SET UP77

rDRATIONAL OF THE EXPERIMENTand the number n of ionizations in the volume is countedd-electrond-electrond-electronprimary trajectoryA narrow parallel particle beam is passing a cylindrical target volume of diameter D, at impact parameter r 8Several years ago we conceived an experiment which allow to MEASURE DIRECTLY some properties of the track structure IN the trajectory of a particle or in its surrounding. 8

rDRATIONAL OF THE EXPERIMENTd-electrond-electrond-electronprimary trajectoryA narrow parallel particle beam is passing a cylindrical target volume of diameter D, at impact parameter r 99

P(G)Peak amplitude h n x GBecause of the small number n of initial electrons, the resolution of measured signal is strongly influenced by gas gain fluctuations. Single electron avalanche varianceHow to count n ?single electron gain distributionIonization measurements based on counting of individual electronsSINGLE IONIZATION (ELECTRON) COUNTINGCounting single electrons1010

MSACDrift ColumnSensitive VolumeA cylinder 3.7 mm in diameter and height filled with 3 mbar of propane (r = 5.47 mg cm-3 at 25), simulates a diameter Dr = 2 mg/cm2.

At density r = 1 g/cm3 : D = 20 nm

ion track

d-electrond-electronsensitivevolume

rSTARTRACK: the device1111

Zoomed view SVSVDRIFT

MSAC1212

13

triggerSSDE FieldsTsLDriftAverage arrival time and std of the time distribution controlled by the drift E field

Gaining factor: longitudinal diffusion

Dispersive factor: transverse diffusionAt low electric fields, the diffusion is symmetric. sL = sT SINGLE IONIZATION (ELECTRON) COUNTING1414STARTRACK: DATA ACQUISITION1515

Waveforms of 16 ms length are sampled at a frequency of 1109s-1 by a PCI analogue to digital acquisition card and stored for off-line analysis. DATA ACQUISITION AND ANALYSIS16

To increase the signal-to-noise ratio, correlation techniques are applied off-line to the raw pulse trails.

Number of pulsesArrival time/nsPeak amplitude/mV39500300.310250400.810800464.6111000328.221113225.21289856.4DATA ANALYSIS: 1. CORRELATION OF RAW DATAIn this technique, each digitized primary electron track is correlated with a typical pulseThe correlation methodPulses above a threshold are counted1717A threshold is applied to correlated pulses.

Number of pulsesArrival time/nsPeak amplitude/mV39500300.310250400.810800464.6111000328.221113225.21289856.4A selective time window is applied in order to reject all the pulse trails that are not consistent with the expected drift time.

DATA ANALYSIS: 2. AMPLITUDE & TIME THRESHOLDSVthr = 40 mVxx11818Background distributions are also measured and unfolding procedure is applied

Measured distribution

Unfolded

background

DATA ANALYSIS: 3. BACKGROUND SUBTRACTION1919Dedicated Monte Carlo code developed (Grosswendt).

Response function included in the Monte Carlo code.

P(G)MSACefficiencyDATA SIMULATIONS20RESULTS OF MEASUREMENTS AND SIMULATIONS2121

rIonization mean free path of the primary ionMean number of electrons set in motion by the primary ion along a path length DRESULTS OF MEASUREMENTS AND SIMULATIONS2222RESULTS OF MEASUREMENTS AND SIMULATIONSCORE REGIONimpact parameter r = 0 nmIonizations produced by the primary particles + ionizations produced by the secondary electrons (d rays)PENUMBRA REGIONimpact parameter r > 0 nmIonizations produced only by d-rays

2323Pn(Q)

M1 = 4.6M1 = 9.2M1 = 0.30M1 = 0.55TRACK CORESparsely ionizing particles, like 20 MeV protons, and densely ionizing particles, like 26.7 MeV 7Li-ions, show markedly different ionization patternsWith decreasing ionization mean free path length of the primary ions the distributions become peaked and shift to larger cluster sizes24Symbols: measurements lines: Monte Carlo simulation24TRACK CORE: same velocity, different Z

Pn(Q)M1 = 0.55M1 = 4.6M1 = 15For given velocity and different Z, M1 is approximately proportional to Z2, in accordance with Bethe theoryM1(2H) = 0.550.55 x 32 = 4.950.55 x 62 = 19.825To improve: Efficiency for large cluster sizes. Symbols: measurements lines: Monte Carlo simulation25

M1= 8.112C, 240 MeVM1= 1512C, 96 MeVlion= 0.26 nm lion= 0.57 nmWith decreasing velocity of the primary ions the distributions shift to larger cluster sizesSlower particles have more time to interact with the target molecules which results in a larger interaction cross section and hence in a smaller ionization mean free path length of primary particles. P nSymbols: measurements lines: Monte Carlo simulationTRACK CORE: same Z, different velocities2626AGREEMENT BETWEEN MEASUREMENTS AND SIMULATIONSDoes it mean that the experimental results are correct?27Validation of the Monte Carlo simulation codeSimulations as a tools to supplement experimental measurementsNO !27RESULTS OF MEASUREMENTS AND SIMULATIONSTRACK STRUCTURE in the PENUMBRA impact parameter r > 20 nmIonizations produced only by the secondary electrons (d rays)2828

P n(r )r = 22 nmr = 30 nmr = 44 nmD = 20 nmCluster-size distributions exhibit the same shape, irrespectively of impact parameterDifferent impact parametersTRACK PENUMBRA: 240 MeV 12C-ions, different r29The shape of the distributions is the same, only they shift to lower occurrence probability of hits for increasing impact parameter. This happens mainly because of the reduction of the solid angle subtended by the target volume at the point of primary impact ionization process, which is the point where secondary electrons are set in motion.29

Pn x (r / D)2240 MeV 12C-ionsr = 22 nmr = 30 nmr = 44 nmD = 20 nmWhen multiplied by the ratio (r / D)2, the distributions form almost a single curve.From the point of view of a nanometer-sized target site, the radiation quality of secondary electrons seems to be the same, only the probability of being hit decreases with increasing r.Different impact parameters comparing different ions30

12C, 240 MeV12C, 96 MeV 1H, 20 MeVP n lion= 0.57 nm lion= 0.26 nm lion= 20.5 nmCluster-size distributions in the penumbra region exhibit the same shape, irrespectively of velocity and Z. lion= 0.55 nm 7Li, 26.7 MeVDifferent Z and velocity?TRACK PENUMBRA: different Z and velocities31

P n / M1 1H, 20 MeV12C, 240 MeV12C, 96 MeV 2H, 16 MeV 7Li, 27 MeV 6Li, 48 MeV 4He, 5.4 MeV lion= 0.55 nm lion= 0.26 nm lion= 20.5 nm lion= 0.55 nm lion= 0.55 nm lion= 9.41 nm lion= 1.05 nmIf divided by M1, the distributions form a single curve, almost independent of radiation quality (charge state and velocity)Different Z and velocityTRACK PENUMBRA after dividing The distributions mainly depend on the number of delta-electrons set in motion, which is here represented by M1. The idea to use M1 is related to the fact that M1 is proportional to D/lamda and is measurable which is not the case in the same experiment for D/lamda.

32

CONCLUSIONSrThe shape of the cluster size distributions is largely independent of r and also of particle type and velocity.The track structure is mainly determined by lion (M1) Cluster-size distributions in nanometre-sized volumes reflect special properties of particle track structure3333240 MeV 12C-IONS HOMOGENEOUS EXTENDED BEAM

Given particles flux~ RV34Here the normalization is to the same dose, shorter radius corresponds to higher density of particles to deliver the same dose.34

Which link to radiobiology?

In radiotherapy a biological sensitive structure is immersed in a homogeneous broad beamR

Measured cluster size distributions can be integrated over the impact parameter to evaluate the track-structure based quality for a nanometer-sized site placed at the centre or at any position of an extended homogeneous beam.

Rfield = 8 mm (45 nm)240 MeV 12C-IONS HOMOGENEOUS EXTENDED BEAM

Here the normalization is to the P(n) integrated over the all beam section (R = 8mm). At shorter distances R, P(n) represents the contribution to the probability distributions due to particles passing at impact parameter less or equal to R. Its clear that the contribution from particles passing at impact parameter larger than about 15 nm is negligible and only increases the relative frequency of zero-events. 35

Challenges: a practical instrument for track structure measurements. a practical instrument (TEPC) for measurements at the nanometer level.

3636