3 <title>Diffraction</title>
4 <link rel="stylesheet" type="text/css" href="pythia.css"/>
5 <link rel="shortcut icon" href="pythia32.gif"/>
13 Diffraction is not well understood, and several alternative approaches
14 have been proposed. Here we follow a fairly conventional Pomeron-based
15 one, in the Ingelman-Schlein spirit [<a href="Bibliography.html" target="page">Ing85</a>],
16 but integrated to make full use of the standard PYTHIA machinery
17 for multiparton interactions, parton showers and hadronization
18 [<a href="Bibliography.html" target="page">Nav10,Cor10a</a>]. This is the approach pioneered in the PomPyt
19 program by Ingelman and collaborators [<a href="Bibliography.html" target="page">Ing97</a>].
22 For ease of use (and of modelling), the Pomeron-specific parts of the
23 generation are subdivided into three sets of parameters that are rather
24 independent of each other:
25 <br/>(i) the total, elastic and diffractive cross sections are
26 parametrized as functions of the CM energy, or can be set by the user
27 to the desired values, see the
28 <a href="TotalCrossSections.html" target="page">Total Cross Sections</a> page;
29 <br/>(ii) once it has been decided to have a diffractive process,
30 a Pomeron flux parametrization is used to pick the mass of the
31 diffractive system(s) and the <i>t</i> of the exchanged Pomeron,
33 <br/>(iii) a diffractive system of a given mass is classified either
34 as low-mass unresolved, which gives a simple low-<i>pT</i> string
35 topology, or as high-mass resolved, for which the full machinery of
36 multiparton interactions and parton showers are applied, making use of
37 <a href="PDFSelection.html" target="page">Pomeron PDFs</a>.
38 <br/>The parameters related to multiparton interactions, parton showers
39 and hadronization are kept the same as for normal nondiffractive events,
40 with only one exception. This may be questioned, especially for the
41 multiparton interactions, but we do not believe that there are currently
42 enough good diffractive data that would allow detailed separate tunes.
45 The above subdivision may not represent the way "physics comes about".
46 For instance, the total diffractive cross section can be viewed as a
47 convolution of a Pomeron flux with a Pomeron-proton total cross section.
48 Since neither of the two is known from first principles there will be
49 a significant amount of ambiguity in the flux factor. The picture is
50 further complicated by the fact that the possibility of simultaneous
51 further multiparton interactions ("cut Pomerons") will screen the rate of
52 diffractive systems. In the end, our set of parameters refers to the
53 effective description that emerges out of these effects, rather than
54 to the underlying "bare" parameters.
57 In the event record the diffractive system in the case of an excited
58 proton is denoted <code>p_diffr</code>, code 9902210, whereas
59 a central diffractive system is denoted <code>rho_diffr</code>,
60 code 9900110. Apart from representing the correct charge and baryon
61 numbers, no deeper meaning should be attributed to the names.
65 As already mentioned above, the total diffractive cross section is fixed
66 by a default energy-dependent parametrization or by the user, see the
67 <a href="TotalCrossSections.html" target="page">Total Cross Sections</a> page.
68 Therefore we do not attribute any significance to the absolute
69 normalization of the Pomeron flux. The choice of Pomeron flux model
70 still will decide on the mass spectrum of diffractive states and the
71 <i>t</i> spectrum of the Pomeron exchange.
73 <p/><code>mode </code><strong> Diffraction:PomFlux </strong>
74 (<code>default = <strong>1</strong></code>; <code>minimum = 1</code>; <code>maximum = 5</code>)<br/>
75 Parametrization of the Pomeron flux <i>f_Pom/p( x_Pom, t)</i>.
76 <br/><code>option </code><strong> 1</strong> : Schuler and Sjöstrand [<a href="Bibliography.html" target="page">Sch94</a>]: based on a
77 critical Pomeron, giving a mass spectrum roughly like <i>dm^2/m^2</i>;
78 a mass-dependent exponential <i>t</i> slope that reduces the rate
79 of low-mass states; partly compensated by a very-low-mass (resonance region)
80 enhancement. Is currently the only one that contains a separate
81 <i>t</i> spectrum for double diffraction (along with MBR) and
82 separate parameters for pion beams.
83 <br/><code>option </code><strong> 2</strong> : Bruni and Ingelman [<a href="Bibliography.html" target="page">Bru93</a>]: also a critical
84 Pomeron giving close to <i>dm^2/m^2</i>, with a <i>t</i> distribution
85 the sum of two exponentials. The original model only covers single
86 diffraction, but is here expanded by analogy to double and central
88 <br/><code>option </code><strong> 3</strong> : a conventional Pomeron description, in the RapGap
89 manual [<a href="Bibliography.html" target="page">Jun95</a>] attributed to Berger et al. and Streng
90 [<a href="Bibliography.html" target="page">Ber87a</a>], but there (and here) with values updated to a
91 supercritical Pomeron with <i>epsilon > 0</i> (see below),
92 which gives a stronger peaking towards low-mass diffractive states,
93 and with a mass-dependent (the <i>alpha'</i> below) exponential
94 <i>t</i> slope. The original model only covers single diffraction,
95 but is here expanded by analogy to double and central diffraction.
97 <br/><code>option </code><strong> 4</strong> : a conventional Pomeron description, attributed to
98 Donnachie and Landshoff [<a href="Bibliography.html" target="page">Don84</a>], again with supercritical Pomeron,
99 with the same two parameters as option 3 above, but this time with a
100 power-law <i>t</i> distribution. The original model only covers single
101 diffraction, but is here expanded by analogy to double and central
103 <br/><code>option </code><strong> 5</strong> : the MBR (Minimum Bias Rockefeller) simulation of
104 (anti)proton-proton interactions [<a href="Bibliography.html" target="page">Cie12</a>]. The event
105 generation follows a renormalized-Regge-theory model, sucessfully tested
106 using CDF data. The simulation includes single and double diffraction,
107 as well as the central diffractive (double-Pomeron exchange) process (106).
108 Only <i>p p</i>, <i>pbar p</i> and <i>p pbar</i> beam combinations
109 are allowed for this option. Several parameters of this model are listed
114 In options 3 and 4 above, the Pomeron Regge trajectory is
117 alpha(t) = 1 + epsilon + alpha' t
119 The <i>epsilon</i> and <i>alpha'</i> parameters can be set
122 <p/><code>parm </code><strong> Diffraction:PomFluxEpsilon </strong>
123 (<code>default = <strong>0.085</strong></code>; <code>minimum = 0.02</code>; <code>maximum = 0.15</code>)<br/>
124 The Pomeron trajectory intercept <i>epsilon</i> above. For technical
125 reasons <i>epsilon > 0</i> is necessary in the current implementation.
127 <p/><code>parm </code><strong> Diffraction:PomFluxAlphaPrime </strong>
128 (<code>default = <strong>0.25</strong></code>; <code>minimum = 0.1</code>; <code>maximum = 0.4</code>)<br/>
129 The Pomeron trajectory slope <i>alpha'</i> above.
132 When option 5 is selected, the following parameters of the MBR model
133 [<a href="Bibliography.html" target="page">Cie12</a>] are used:
135 <p/><code>parm </code><strong> Diffraction:MBRepsilon </strong>
136 (<code>default = <strong>0.104</strong></code>; <code>minimum = 0.02</code>; <code>maximum = 0.15</code>)<br/>
137 <p/><code>parm </code><strong> Diffraction:MBRalpha </strong>
138 (<code>default = <strong>0.25</strong></code>; <code>minimum = 0.1</code>; <code>maximum = 0.4</code>)<br/>
139 the parameters of the Pomeron trajectory.
141 <p/><code>parm </code><strong> Diffraction:MBRbeta0 </strong>
142 (<code>default = <strong>6.566</strong></code>; <code>minimum = 0.0</code>; <code>maximum = 10.0</code>)<br/>
143 <p/><code>parm </code><strong> Diffraction:MBRsigma0 </strong>
144 (<code>default = <strong>2.82</strong></code>; <code>minimum = 0.0</code>; <code>maximum = 5.0</code>)<br/>
145 the Pomeron-proton coupling, and the total Pomeron-proton cross section.
147 <p/><code>parm </code><strong> Diffraction:MBRm2Min </strong>
148 (<code>default = <strong>1.5</strong></code>; <code>minimum = 0.0</code>; <code>maximum = 3.0</code>)<br/>
149 the lowest value of the mass squared of the dissociated system.
151 <p/><code>parm </code><strong> Diffraction:MBRdyminSDflux </strong>
152 (<code>default = <strong>2.3</strong></code>; <code>minimum = 0.0</code>; <code>maximum = 5.0</code>)<br/>
153 <p/><code>parm </code><strong> Diffraction:MBRdyminDDflux </strong>
154 (<code>default = <strong>2.3</strong></code>; <code>minimum = 0.0</code>; <code>maximum = 5.0</code>)<br/>
155 <p/><code>parm </code><strong> Diffraction:MBRdyminCDflux </strong>
156 (<code>default = <strong>2.3</strong></code>; <code>minimum = 0.0</code>; <code>maximum = 5.0</code>)<br/>
157 the minimum width of the rapidity gap used in the calculation of
158 <i>Ngap(s)</i> (flux renormalization).
160 <p/><code>parm </code><strong> Diffraction:MBRdyminSD </strong>
161 (<code>default = <strong>2.0</strong></code>; <code>minimum = 0.0</code>; <code>maximum = 5.0</code>)<br/>
162 <p/><code>parm </code><strong> Diffraction:MBRdyminDD </strong>
163 (<code>default = <strong>2.0</strong></code>; <code>minimum = 0.0</code>; <code>maximum = 5.0</code>)<br/>
164 <p/><code>parm </code><strong> Diffraction:MBRdyminCD </strong>
165 (<code>default = <strong>2.0</strong></code>; <code>minimum = 0.0</code>; <code>maximum = 5.0</code>)<br/>
166 the minimum width of the rapidity gap used in the calculation of cross
167 sections, i.e. the parameter <i>dy_S</i>, which suppresses the cross
168 section at low <i>dy</i> (non-diffractive region).
170 <p/><code>parm </code><strong> Diffraction:MBRdyminSigSD </strong>
171 (<code>default = <strong>0.5</strong></code>; <code>minimum = 0.001</code>; <code>maximum = 5.0</code>)<br/>
172 <p/><code>parm </code><strong> Diffraction:MBRdyminSigDD </strong>
173 (<code>default = <strong>0.5</strong></code>; <code>minimum = 0.001</code>; <code>maximum = 5.0</code>)<br/>
174 <p/><code>parm </code><strong> Diffraction:MBRdyminSigCD </strong>
175 (<code>default = <strong>0.5</strong></code>; <code>minimum = 0.001</code>; <code>maximum = 5.0</code>)<br/>
176 the parameter <i>sigma_S</i>, used for the cross section suppression at
177 low <i>dy</i> (non-diffractive region).
179 <h3>Separation into low and high masses</h3>
181 Preferably one would want to have a perturbative picture of the
182 dynamics of Pomeron-proton collisions, like multiparton interactions
183 provide for proton-proton ones. However, while PYTHIA by default
184 will only allow collisions with a CM energy above 10 GeV, the
185 mass spectrum of diffractive systems will stretch to down to
186 the order of 1.2 GeV. It would not be feasible to attempt a
187 perturbative description there. Therefore we do offer a simpler
188 low-mass description, with only longitudinally stretched strings,
189 with a gradual switch-over to the perturbative picture for higher
190 masses. The probability for the latter picture is parametrized as
192 P_pert = P_max ( 1 - exp( (m_diffr - m_min) / m_width ) )
194 which vanishes for the diffractive system mass
195 <i>m_diffr < m_min</i>, and is <i>1 - 1/e = 0.632</i> for
196 <i>m_diffr = m_min + m_width</i>, assuming <i>P_max = 1</i>.
198 <p/><code>parm </code><strong> Diffraction:mMinPert </strong>
199 (<code>default = <strong>10.</strong></code>; <code>minimum = 5.</code>)<br/>
200 The abovementioned threshold mass <i>m_min</i> for phasing in a
201 perturbative treatment. If you put this parameter to be bigger than
202 the CM energy then there will be no perturbative description at all,
203 but only the older low-<i>pt</i> description.
206 <p/><code>parm </code><strong> Diffraction:mWidthPert </strong>
207 (<code>default = <strong>10.</strong></code>; <code>minimum = 0.</code>)<br/>
208 The abovementioned threshold width <i>m_width.</i>
211 <p/><code>parm </code><strong> Diffraction:probMaxPert </strong>
212 (<code>default = <strong>1.</strong></code>; <code>minimum = 0.</code>; <code>maximum = 1.</code>)<br/>
213 The abovementioned maximum probability <i>P_max.</i>. Would
214 normally be assumed to be unity, but a somewhat lower value could
215 be used to represent a small nonperturbative component also at
216 high diffractive masses.
219 <h3>Low-mass diffraction</h3>
221 When an incoming hadron beam is diffractively excited, it is modeled
222 as if either a valence quark or a gluon is kicked out from the hadron.
223 In the former case this produces a simple string to the leftover
224 remnant, in the latter it gives a hairpin arrangement where a string
225 is stretched from one quark in the remnant, via the gluon, back to the
226 rest of the remnant. The latter ought to dominate at higher mass of
227 the diffractive system. Therefore an approximate behaviour like
233 <p/><code>parm </code><strong> Diffraction:pickQuarkNorm </strong>
234 (<code>default = <strong>5.0</strong></code>; <code>minimum = 0.</code>)<br/>
235 The abovementioned normalization <i>N</i> for the relative quark
236 rate in diffractive systems.
239 <p/><code>parm </code><strong> Diffraction:pickQuarkPower </strong>
240 (<code>default = <strong>1.0</strong></code>)<br/>
241 The abovementioned mass-dependence power <i>p</i> for the relative
242 quark rate in diffractive systems.
246 When a gluon is kicked out from the hadron, the longitudinal momentum
247 sharing between the the two remnant partons is determined by the
248 same parameters as above. It is plausible that the primordial
249 <i>kT</i> may be lower than in perturbative processes, however:
251 <p/><code>parm </code><strong> Diffraction:primKTwidth </strong>
252 (<code>default = <strong>0.5</strong></code>; <code>minimum = 0.</code>)<br/>
253 The width of Gaussian distributions in <i>p_x</i> and <i>p_y</i>
254 separately that is assigned as a primordial <i>kT</i> to the two
255 beam remnants when a gluon is kicked out of a diffractive system.
258 <p/><code>parm </code><strong> Diffraction:largeMassSuppress </strong>
259 (<code>default = <strong>2.</strong></code>; <code>minimum = 0.</code>)<br/>
260 The choice of longitudinal and transverse structure of a diffractive
261 beam remnant for a kicked-out gluon implies a remnant mass
262 <i>m_rem</i> distribution (i.e. quark plus diquark invariant mass
263 for a baryon beam) that knows no bounds. A suppression like
264 <i>(1 - m_rem^2 / m_diff^2)^p</i> is therefore introduced, where
265 <i>p</i> is the <code>diffLargeMassSuppress</code> parameter.
268 <h3>High-mass diffraction</h3>
270 The perturbative description need to use parton densities of the
271 Pomeron. The options are described in the page on
272 <a href="PDFSelection.html" target="page">PDF Selection</a>. The standard
273 perturbative multiparton interactions framework then provides
274 cross sections for parton-parton interactions. In order to
275 turn these cross section into probabilities one also needs an
276 ansatz for the Pomeron-proton total cross section. In the literature
277 one often finds low numbers for this, of the order of 2 mb.
278 These, if taken at face value, would give way too much activity
279 per event. There are ways to tame this, e.g. by a larger <i>pT0</i>
280 than in the normal pp framework. Actually, there are many reasons
281 to use a completely different set of parameters for MPI in
282 diffraction than in pp collisions, especially with respect to the
283 impact-parameter picture, see below. A lower number in some frameworks
284 could alternatively be regarded as a consequence of screening, with
285 a larger "bare" number.
288 For now, however, an attempt at the most general solution would
289 carry too far, and instead we patch up the problem by using a
290 larger Pomeron-proton total cross section, such that average
291 activity makes more sense. This should be viewed as the main
292 tunable parameter in the description of high-mass diffraction.
293 It is to be fitted to diffractive event-shape data such as the average
294 charged multiplicity. It would be very closely tied to the choice of
295 Pomeron PDF; we remind that some of these add up to less than unit
296 momentum sum in the Pomeron, a choice that also affect the value
297 one ends up with. Furthermore, like with hadronic cross sections,
298 it is quite plausible that the Pomeron-proton cross section increases
299 with energy, so we have allowed for a powerlike dependence on the
302 <p/><code>parm </code><strong> Diffraction:sigmaRefPomP </strong>
303 (<code>default = <strong>10.</strong></code>; <code>minimum = 2.</code>; <code>maximum = 40.</code>)<br/>
304 The assumed Pomeron-proton effective cross section, as used for
305 multiparton interactions in diffractive systems. If this cross section
306 is made to depend on the mass of the diffractive system then the above
307 value refers to the cross section at the reference scale, and
309 sigma_PomP(m) = sigma_PomP(m_ref) * (m / m_ref)^p
311 where <i>m</i> is the mass of the diffractive system, <i>m_ref</i>
312 is the reference mass scale <code>Diffraction:mRefPomP</code> below and
313 <i>p</i> is the mass-dependence power <code>Diffraction:mPowPomP</code>.
314 Note that a larger cross section value gives less MPI activity per event.
315 There is no point in making the cross section too big, however, since
316 then <i>pT0</i> will be adjusted downwards to ensure that the
317 integrated perturbative cross section stays above this assumed total
318 cross section. (The requirement of at least one perturbative interaction
322 <p/><code>parm </code><strong> Diffraction:mRefPomP </strong>
323 (<code>default = <strong>100.0</strong></code>; <code>minimum = 1.</code>)<br/>
324 The <i>mRef</i> reference mass scale introduced above.
327 <p/><code>parm </code><strong> Diffraction:mPowPomP </strong>
328 (<code>default = <strong>0.0</strong></code>; <code>minimum = 0.0</code>; <code>maximum = 0.5</code>)<br/>
329 The <i>p</i> mass rescaling pace introduced above.
333 Also note that, even for a fixed CM energy of events, the diffractive
334 subsystem will range from the abovementioned threshold mass
335 <i>m_min</i> to the full CM energy, with a variation of parameters
336 such as <i>pT0</i> along this mass range. Therefore multiparton
337 interactions are initialized for a few different diffractive masses,
338 currently five, and all relevant parameters are interpolated between
339 them to obtain the behaviour at a specific diffractive mass.
340 Furthermore, <i>A B ->X B</i> and <i>A B ->A X</i> are
341 initialized separately, to allow for different beams or PDF's on the
342 two sides. These two aspects mean that initialization of MPI is
343 appreciably slower when perturbative high-mass diffraction is allowed.
346 Diffraction tends to be peripheral, i.e. occur at intermediate impact
347 parameter for the two protons. That aspect is implicit in the selection
348 of diffractive cross section. For the simulation of the Pomeron-proton
349 subcollision it is the impact-parameter distribution of that particular
350 subsystem that should rather be modelled. That is, it also involves
351 the transverse coordinate space of a Pomeron wavefunction. The outcome
352 of the convolution therefore could be a different shape than for
353 nondiffractive events. For simplicity we allow the same kind of
354 options as for nondiffractive events, except that the
355 <code>bProfile = 4</code> option for now is not implemented.
357 <p/><code>mode </code><strong> Diffraction:bProfile </strong>
358 (<code>default = <strong>1</strong></code>; <code>minimum = 0</code>; <code>maximum = 3</code>)<br/>
359 Choice of impact parameter profile for the incoming hadron beams.
360 <br/><code>option </code><strong> 0</strong> : no impact parameter dependence at all.
361 <br/><code>option </code><strong> 1</strong> : a simple Gaussian matter distribution;
363 <br/><code>option </code><strong> 2</strong> : a double Gaussian matter distribution,
364 with the two free parameters <i>coreRadius</i> and
366 <br/><code>option </code><strong> 3</strong> : an overlap function, i.e. the convolution of
367 the matter distributions of the two incoming hadrons, of the form
368 <i>exp(- b^expPow)</i>, where <i>expPow</i> is a free
372 <p/><code>parm </code><strong> Diffraction:coreRadius </strong>
373 (<code>default = <strong>0.4</strong></code>; <code>minimum = 0.1</code>; <code>maximum = 1.</code>)<br/>
374 When assuming a double Gaussian matter profile, <i>bProfile = 2</i>,
375 the inner core is assumed to have a radius that is a factor
376 <i>coreRadius</i> smaller than the rest.
379 <p/><code>parm </code><strong> Diffraction:coreFraction </strong>
380 (<code>default = <strong>0.5</strong></code>; <code>minimum = 0.</code>; <code>maximum = 1.</code>)<br/>
381 When assuming a double Gaussian matter profile, <i>bProfile = 2</i>,
382 the inner core is assumed to have a fraction <i>coreFraction</i>
383 of the matter content of the hadron.
386 <p/><code>parm </code><strong> Diffraction:expPow </strong>
387 (<code>default = <strong>1.</strong></code>; <code>minimum = 0.4</code>; <code>maximum = 10.</code>)<br/>
388 When <i>bProfile = 3</i> it gives the power of the assumed overlap
389 shape <i>exp(- b^expPow)</i>. Default corresponds to a simple
390 exponential drop, which is not too dissimilar from the overlap
391 obtained with the standard double Gaussian parameters. For
392 <i>expPow = 2</i> we reduce to the simple Gaussian, <i>bProfile = 1</i>,
393 and for <i>expPow -> infinity</i> to no impact parameter dependence
394 at all, <i>bProfile = 0</i>. For small <i>expPow</i> the program
395 becomes slow and unstable, so the min limit must be respected.
401 <!-- Copyright (C) 2012 Torbjorn Sjostrand -->