source: HiSusy/trunk/Pythia8/pythia8170/htmldoc/Fragmentation.html @ 1

Last change on this file since 1 was 1, checked in by zerwas, 11 years ago

first import of structure, PYTHIA8 and DELPHES

File size: 15.3 KB
Line 
1<html>
2<head>
3<title>Fragmentation</title>
4<link rel="stylesheet" type="text/css" href="pythia.css"/>
5<link rel="shortcut icon" href="pythia32.gif"/>
6</head>
7<body>
8
9<h2>Fragmentation</h2>
10
11Fragmentation in PYTHIA is based on the Lund string model
12[<a href="Bibliography.html" target="page">And83, Sjo84</a>]. Several different aspects are involved in
13the physics description, which  here therefore is split accordingly.
14This also, at least partly, reflect the set of classes involved in
15the fragmentation machinery.
16
17<p/>
18The variables collected here have a very wide span of usefulness.
19Some would be central in any hadronization tuning exercise, others
20should not be touched except by experts.
21
22<p/>
23The fragmentation flavour-choice machinery is also used in a few
24other places of the program, notably particle decays, and is thus
25described on the separate <a href="FlavourSelection.html" target="page">Flavour
26Selection</a> page.
27
28<h3>Fragmentation functions</h3>
29
30The <code>StringZ</code> class handles the choice of longitudinal
31lightcone fraction <i>z</i> according to one of two possible
32shape sets.
33
34<p/>
35The Lund symmetric fragmentation function [<a href="Bibliography.html" target="page">And83</a>] is the
36only alternative for light quarks. It is of the form
37<br/><i> 
38    f(z) = (1/z) * (1-z)^a * exp(-b m_T^2 / z)
39</i><br/>
40with the two main free parameters <i>a</i> and <i>b</i> to be
41tuned to data. They are stored in
42
43<p/><code>parm&nbsp; </code><strong> StringZ:aLund &nbsp;</strong> 
44 (<code>default = <strong>0.3</strong></code>; <code>minimum = 0.0</code>; <code>maximum = 2.0</code>)<br/>
45The <i>a</i> parameter of the Lund symmetric fragmentation function.
46 
47
48<p/><code>parm&nbsp; </code><strong> StringZ:bLund &nbsp;</strong> 
49 (<code>default = <strong>0.8</strong></code>; <code>minimum = 0.2</code>; <code>maximum = 2.0</code>)<br/>
50The <i>b</i> parameter of the Lund symmetric fragmentation function.
51 
52
53<p/>
54In principle, each flavour can have a different <i>a</i>. Then,
55for going from an old flavour <i>i</i> to a new <i>j</i> one
56the shape is
57<br/><i> 
58    f(z) = (1/z) * z^{a_i} * ((1-z)/z)^{a_j} * exp(-b * m_T^2 / z)
59</i><br/>
60This is only implemented for diquarks relative to normal quarks:
61
62<p/><code>parm&nbsp; </code><strong> StringZ:aExtraDiquark &nbsp;</strong> 
63 (<code>default = <strong>0.5</strong></code>; <code>minimum = 0.0</code>; <code>maximum = 2.0</code>)<br/>
64allows a larger <i>a</i> for diquarks, with total
65<i>a = aLund + aExtraDiquark</i>.
66 
67
68<p/>
69Finally, the Bowler modification [<a href="Bibliography.html" target="page">Bow81</a>] introduces an extra
70factor
71<br/><i>
72    1/z^{r_Q * b * m_Q^2}
73</i><br/>
74for heavy quarks. To keep some flexibility, a multiplicative factor
75<i>r_Q</i> is introduced, which ought to be unity (provided that
76quark masses were uniquely defined) but can be set in
77
78<p/><code>parm&nbsp; </code><strong> StringZ:rFactC &nbsp;</strong> 
79 (<code>default = <strong>1.0</strong></code>; <code>minimum = 0.0</code>; <code>maximum = 2.0</code>)<br/>
80<i>r_c</i>, i.e. the above parameter for <i>c</i> quarks.
81 
82
83<p/><code>parm&nbsp; </code><strong> StringZ:rFactB &nbsp;</strong> 
84 (<code>default = <strong>0.67</strong></code>; <code>minimum = 0.0</code>; <code>maximum = 2.0</code>)<br/>
85<i>r_b</i>, i.e. the above parameter for <i>b</i> quarks.
86 
87
88<p/><code>parm&nbsp; </code><strong> StringZ:rFactH &nbsp;</strong> 
89 (<code>default = <strong>1.0</strong></code>; <code>minimum = 0.0</code>; <code>maximum = 2.0</code>)<br/>
90<i>r_h</i>, i.e. the above parameter for heavier hypothetical quarks,
91or in general any new coloured particle long-lived enough to hadronize.
92 
93
94<p/>
95As an alternative, it is possible to switch over to the
96Peterson/SLAC formula [<a href="Bibliography.html" target="page">Pet83</a>]
97<br/><i>
98     f(z) = 1 / ( z * (1 - 1/z - epsilon/(1-z))^2 )
99</i><br/>
100for charm, bottom and heavier (defined as above) by the three flags
101
102<p/><code>flag&nbsp; </code><strong> StringZ:usePetersonC &nbsp;</strong> 
103 (<code>default = <strong>off</strong></code>)<br/>
104use Peterson for <i>c</i> quarks.
105 
106
107<p/><code>flag&nbsp; </code><strong> StringZ:usePetersonB &nbsp;</strong> 
108 (<code>default = <strong>off</strong></code>)<br/>
109use Peterson for <i>b</i> quarks.
110 
111
112<p/><code>flag&nbsp; </code><strong> StringZ:usePetersonH &nbsp;</strong> 
113 (<code>default = <strong>off</strong></code>)<br/>
114use Peterson for hypothetical heavier quarks.
115 
116
117<p/>
118When switched on, the corresponding epsilon values are chosen to be
119
120<p/><code>parm&nbsp; </code><strong> StringZ:epsilonC &nbsp;</strong> 
121 (<code>default = <strong>0.05</strong></code>; <code>minimum = 0.01</code>; <code>maximum = 0.25</code>)<br/>
122<i>epsilon_c</i>, i.e. the above parameter for <i>c</i> quarks.
123 
124
125<p/><code>parm&nbsp; </code><strong> StringZ:epsilonB &nbsp;</strong> 
126 (<code>default = <strong>0.005</strong></code>; <code>minimum = 0.001</code>; <code>maximum = 0.025</code>)<br/>
127<i>epsilon_b</i>, i.e. the above parameter for <i>b</i> quarks.
128 
129
130<p/><code>parm&nbsp; </code><strong> StringZ:epsilonH &nbsp;</strong> 
131 (<code>default = <strong>0.005</strong></code>; <code>minimum = 0.0001</code>; <code>maximum = 0.25</code>)<br/>
132<i>epsilon_h</i>, i.e. the above parameter for hypothetical heavier
133quarks, normalized to the case where <i>m_h = m_b</i>. The actually
134used parameter is then <i>epsilon = epsilon_h * (m_b^2 / m_h^2)</i>.
135This allows a sensible scaling to a particle with an unknown higher
136mass without the need for a user intervention.
137 
138
139<h3>Fragmentation <i>pT</i></h3>
140
141The <code>StringPT</code> class handles the choice of fragmentation
142<i>pT</i>. At each string breaking the quark and antiquark of the pair are
143supposed to receive opposite and compensating <i>pT</i> kicks according
144to a Gaussian distribution in <i>p_x</i> and <i>p_y</i> separately.
145Call <i>sigma_q</i> the width of the <i>p_x</i> and <i>p_y</i> 
146distributions separately, i.e.
147<br/><i>
148    d(Prob) = exp( -(p_x^2 + p_y^2) / 2 sigma_q^2).
149</i><br/>
150Then the total squared width is
151<br/><i>
152    &lt;pT^2> = &lt;p_x^2> +  &lt;p_y^2> = 2 sigma_q^2 = sigma^2.
153</i><br/>
154It is this latter number that is stored in
155
156<p/><code>parm&nbsp; </code><strong> StringPT:sigma &nbsp;</strong> 
157 (<code>default = <strong>0.304</strong></code>; <code>minimum = 0.0</code>; <code>maximum = 1.0</code>)<br/>
158the width <i>sigma</i> in the fragmentation process.
159 
160
161<p/>
162Since a normal hadron receives <i>pT</i> contributions for two string
163breakings, it has a <i>&lt;p_x^2>_had = &lt;p_y^2>_had = sigma^2</i>,
164and thus <i>&lt;pT^2>_had = 2 sigma^2</i>
165
166<p/>
167Some studies on isolated particles at LEP has indicated the need for
168a slightly enhanced rate in the high-<i>pT</i> tail of the above
169distribution. This would have to be reviewed in the context of a
170complete retune of parton showers and hadronization, but for the
171moment we stay with the current recipe, to boost the above <i>pT</i> 
172by a factor <i>enhancedWidth</i> for a small fraction
173<i>enhancedFraction</i> of the breakups, where
174
175<p/><code>parm&nbsp; </code><strong> StringPT:enhancedFraction &nbsp;</strong> 
176 (<code>default = <strong>0.01</strong></code>; <code>minimum = 0.0</code>; <code>maximum = 1.</code>)<br/>
177<i>enhancedFraction</i>,the fraction of string breaks with enhanced
178width.
179 
180
181<p/><code>parm&nbsp; </code><strong> StringPT:enhancedWidth &nbsp;</strong> 
182 (<code>default = <strong>2.0</strong></code>; <code>minimum = 1.0</code>; <code>maximum = 10.0</code>)<br/>
183<i>enhancedWidth</i>,the enhancement of the width in this fraction.
184 
185
186<h3>Jet joining procedure</h3>
187
188String fragmentation is carried out iteratively from both string ends
189inwards, which means that the two chains of hadrons have to be joined up
190somewhere in the middle of the event. This joining is described by
191parameters that in principle follows from the standard fragmentation
192parameters, but in a way too complicated to parametrize. The dependence
193is rather mild, however, so for a sensible range of variation the
194parameters in this section should not be touched.
195
196<p/><code>parm&nbsp; </code><strong> StringFragmentation:stopMass &nbsp;</strong> 
197 (<code>default = <strong>1.0</strong></code>; <code>minimum = 0.0</code>; <code>maximum = 2.0</code>)<br/>
198Is used to define a <i>W_min = m_q1 + m_q2 + stopMass</i>,
199where <i>m_q1</i> and <i>m_q2</i> are the masses of the two
200current endpoint quarks or diquarks.
201 
202
203<p/><code>parm&nbsp; </code><strong> StringFragmentation:stopNewFlav &nbsp;</strong> 
204 (<code>default = <strong>2.0</strong></code>; <code>minimum = 0.0</code>; <code>maximum = 2.0</code>)<br/>
205Add to <i>W_min</i> an amount <i>stopNewFlav * m_q_last</i>,
206where <i>q_last</i> is the last <i>q qbar</i> pair produced
207between the final two hadrons.
208 
209
210<p/><code>parm&nbsp; </code><strong> StringFragmentation:stopSmear &nbsp;</strong> 
211 (<code>default = <strong>0.2</strong></code>; <code>minimum = 0.0</code>; <code>maximum = 0.5</code>)<br/>
212The <i>W_min</i> above is then smeared uniformly in the range
213<i>W_min_smeared = W_min * [ 1 - stopSmear, 1 + stopSmear ]</i>.
214 
215
216<p/>
217This <i>W_min_smeared</i> is then compared with the current remaining
218<i>W_transverse</i> to determine if there is energy left for further
219particle production. If not, i.e. if
220<i>W_transverse &lt; W_min_smeared</i>, the final two particles are
221produced from what is currently left, if possible. (If not, the
222fragmentation process is started over.)
223
224<h3>Simplifying systems</h3>
225
226There are a few situations when it is meaningful to simplify the
227original task, one way or another.
228
229<p/><code>parm&nbsp; </code><strong> HadronLevel:mStringMin &nbsp;</strong> 
230 (<code>default = <strong>1.</strong></code>; <code>minimum = 0.5</code>; <code>maximum = 1.5</code>)<br/>
231Decides whether a partonic system should be considered as a normal
232string or a ministring, the latter only producing one or two primary
233hadrons. The system mass should be above <i>mStringMin</i> plus the
234sum of quark/diquark constituent masses for a normal string description,
235else the ministring scenario is used.
236 
237
238<p/><code>parm&nbsp; </code><strong> FragmentationSystems:mJoin &nbsp;</strong> 
239 (<code>default = <strong>0.3</strong></code>; <code>minimum = 0.2</code>; <code>maximum = 1.</code>)<br/>
240When two colour-connected partons are very nearby, with at least
241one being a gluon, they can be joined into one, to avoid technical
242problems of very small string regions. The requirement for joining is
243that the invariant mass of the pair is below <i>mJoin</i>, where a
244gluon only counts with half its momentum, i.e. with its contribution
245to the string region under consideration. (Note that, for technical
246reasons, the 0.2 GeV lower limit is de facto hardcoded.)
247 
248
249<p/><code>parm&nbsp; </code><strong> FragmentationSystems:mJoinJunction &nbsp;</strong> 
250 (<code>default = <strong>1.0</strong></code>; <code>minimum = 0.5</code>; <code>maximum = 2.</code>)<br/>
251When the invariant mass of two of the quarks in a three-quark junction
252string system becomes too small, the system is simplified to a
253quark-diquark simple string. The requirement for this simplification
254is that the diquark mass, minus the two quark masses, falls below
255<i>mJoinJunction</i>. Gluons on the string between the junction and
256the respective quark, if any, are counted as part of the quark
257four-momentum. Those on the two combined legs are clustered with the
258diquark when it is formed.
259 
260
261<h3>Ministrings</h3>
262
263The <code>MiniStringFragmentation</code> machinery is only used when a
264string system has so small invariant mass that normal string fragmentation
265is difficult/impossible. Instead one or two particles are produced,
266in the former case shuffling energy-momentum relative to another
267colour singlet system in the event, while preserving the invariant
268mass of that system. With one exception parameters are the same as
269defined for normal string fragmentation, to the extent that they are
270at all applicable in this case.
271
272A discussion of the relevant physics is found in [<a href="Bibliography.html" target="page">Nor00</a>].
273The current implementation does not completely abide to the scheme
274presented there, however, but has in part been simplified. (In part
275for greater clarity, in part since the class is not quite finished yet.)
276
277<p/><code>mode&nbsp; </code><strong> MiniStringFragmentation:nTry &nbsp;</strong> 
278 (<code>default = <strong>2</strong></code>; <code>minimum = 1</code>; <code>maximum = 10</code>)<br/>
279Whenever the machinery is called, first this many attempts are made
280to pick two hadrons that the system fragments to. If the hadrons are
281too massive the attempt will fail, but a new subsequent try could
282involve other flavour and hadrons and thus still succeed.
283After <i>nTry</i> attempts, instead an attempt is made to produce a
284single hadron from the system. Should also this fail, some further
285attempts at obtaining two hadrons will be made before eventually
286giving up.
287 
288
289<h3>Junction treatment</h3>
290
291A junction topology corresponds to an Y arrangement of strings
292i.e. where three string pieces have to be joined up in a junction.
293Such topologies can arise if several valence quarks are kicked out
294from a proton beam, or in baryon-number-violating SUSY decays.
295Special attention is necessary to handle the region just around
296the junction, where the baryon number topologically is located.
297The junction fragmentation scheme is described in [<a href="Bibliography.html" target="page">Sjo03</a>].
298The parameters in this section should not be touched except by experts.
299
300<p/><code>parm&nbsp; </code><strong> StringFragmentation:eNormJunction &nbsp;</strong> 
301 (<code>default = <strong>2.0</strong></code>; <code>minimum = 0.5</code>; <code>maximum = 10</code>)<br/>
302Used to find the effective rest frame of the junction, which is
303complicated when the three string legs may contain additional
304gluons between the junction and the endpoint. To this end,
305a pull is defined as a weighed sum of the momenta on each leg,
306where the weight is <i>exp(- eSum / eNormJunction)</i>, with
307<i>eSum</i> the summed energy of all partons closer to the junction
308than the currently considered one (in the junction rest frame).
309Should in principle be (close to) <i>sqrt((1 + a) / b)</i>, with
310<i>a</i> and <i>b</i> the parameters of the Lund symmetric
311fragmentation function.
312 
313
314<p/><code>parm&nbsp; </code><strong> StringFragmentation:eBothLeftJunction &nbsp;</strong> 
315 (<code>default = <strong>1.0</strong></code>; <code>minimum = 0.5</code>)<br/>
316Retry (up to 10 times) when the first two considered strings in to a
317junction both have a remaining energy (in the junction rest frame)
318above this number.
319 
320
321<p/><code>parm&nbsp; </code><strong> StringFragmentation:eMaxLeftJunction &nbsp;</strong> 
322 (<code>default = <strong>10.0</strong></code>; <code>minimum = 0.</code>)<br/>
323Retry (up to 10 times) when the first two considered strings in to a
324junction has a highest remaining energy (in the junction rest frame)
325above a random energy evenly distributed between
326<i>eBothLeftJunction</i> and
327<i>eBothLeftJunction + eMaxLeftJunction</i> 
328(drawn anew for each test).
329 
330
331<p/><code>parm&nbsp; </code><strong> StringFragmentation:eMinLeftJunction &nbsp;</strong> 
332 (<code>default = <strong>0.2</strong></code>; <code>minimum = 0.</code>)<br/>
333Retry (up to 10 times) when the invariant mass-squared of the final leg
334and the leftover momentum of the first two treated legs falls below
335<i>eMinLeftJunction</i> times the energy of the final leg (in the
336junction rest frame).
337 
338
339</body>
340</html>
341
342<!-- Copyright (C) 2012 Torbjorn Sjostrand -->
343
Note: See TracBrowser for help on using the repository browser.