See CDF-notes 5288 (physics analisys) and 5575 (complementary studies and efficiency corrections).
 

In this study we try to characterize soft interactions (whatever they are) and to find any peculiar behavior which
exhibit specific properties that may help us to understand them.
We start working with a sample of events which are collected without requiring any particular "a priori" condition
("minimum bias" events). Out of  minimum-bias events we select the "soft" ones by applying the only selection
criterium that we know to be valid for them: we require that they do not show up as jets (jets are spreads of collimated
fast particles that are typical of head-on parton collisions). In our particular definition of "soft" events we include all
events with no jets of energy greater than 1.1 GeV. This number is suggested by the detector ability to measure a jet
energy, but we show that the results do not critically depend on this particular  value.

Finally we study the properties of selected soft events as compared to all the others. In particular we analyze:

• the charged multiplicity distribution. This is the probability for an event to produce a given number of final particles in the central region.
• the  transverse momentum distribution.  This is the probability for particle produced in the central region to have a given transverse momentum
• the fluctuation (from event to event) of the event average transverse momentum. This is an indication of what  is the spread of the mean energy released transversally to the colliding axis.

This is an unexpected and new result, and the analysis is the first one of this kind ever published.
 

 Part of the plots blessed for this analisys:

 1. primary charged particles multiplicity distributions, in KNO form (MB)
 2. primary charged particles multiplicity distributions, in KNO form (soft)
 3. primary charged particles multiplcity distributions, in KNO form (hard)
 4. average charged particles pT vs multiplcity (MB)
 5. average charged particles pT vs multplicity (soft)
 6. average charged particles pT vs multiplcity (hard)
 

See CDF-notes 6075 and 6043 (pick latest versions).
 

Besides the lighter quarks u and d , strangeness production is the only component of low-pT
multiparticle interactions to be statistically significant and experimentally accessible with a
Minimum-Bias trigger. It is also a valid probe to investigate the transition of soft hadron
interactions to the high-pT QCD perturbative region.

This paper describes an analysis of k and Lambda (V0) production in pp interactions. It is part of
a low-pT multiparticle production systematic study structured in comparative analyses of
statistical distributions and particle correlations at different c.m.s. energies.
Specific emphasis is given to the dependence of the particle correlations on charged multiplicity
and to the ``hardness'' of the interaction.

In analogy with the paper described above the whole analysis was repeated on two different types
of events that have been selected by dividing MB data into soft and hard sub-samples.

Inclusive distributions of multiplicity and transverse momentum of K and Lambda are presented
first. The high statistics of the data sample collected at sqrt(s)=1800 and 630 GeV and  accurate
efficiency corrections, allow to extend the range of these measurements and their precision with
respect to previous ones.

Studies of the dependence of the average pT of V0 and of their mean number on the event charged
multiplicity are also presented. It is not possible to enphasize any difference of the <pT> dependence
with multiplicity at the two energies, even in the full MB sample. This is even more so for the soft
and hard sub-samples. Nevertheless the behavior of the three sub-samples is clearly different.

The kinematic selection of K and Lambdas (described in note 6075)

1. look for opposite sign CTC track pairs converging to a secondary vertex;
2. fit secondary vertex with both K and L0 hypotheses and keep best fit;
3. secondary vertex 3-C fit must have probability >5%;
4. fitted mass must be within 3 sigmas from k/L0 mass;
5. vertex displacement projected in x-y plane (Lxy) >1 cm;
6. decay products in pT>0.300 GeV/c and |eta|<1.5;
7. | Z(V0) - Z(event) | < 6 cm;
8. d0(V0) < 0.7 cm;
9. pT(V0) > 0.4 GeV/c;
10. | eta(V0) | < 1.0

Includes efficiency + "fakes" + acceptance + contamination.
Efficiency: is computed in two ways: with full MC generation/simulation/reconstruction and
                   by embedding fake k/L0 into real min-bias events;
Fakes: we mean by this fake associations of secondary tracks, but also V0 which are
           reconstructed outside of our defined limits for the efficiency and also the
           contamination of K in the L0 sample and viceversa;
Acceptance: originated by our fiducial cuts in Lxy  and pT of the V0 decay products
Overall correction is computed as C = (1 - Fakes)/(Efficiency x Acceptance)

Blessed plots:

1. invariant mass distributions of pi-pi and p-pi pairs (MB, 1800 GeV)
  Invariant mass distribution of the decay products of k and Lambdas after kinematical selection.
  No correction is applied.

2. lifetime distribution of k, raw and corrected (MB, 1800 GeV)
  The lifetime distribution of k is shown before and after effciency correction.
  The mean value is (0.881+-0.006)E-10 s

3. lifetime distribution for Lambda, raw and corrected (MB, 1800 GeV)
  The lifetime distribution of k is shown before and after effciency correction.
  Te mean value is (2.62+-0.08)E-10 s

4. distribution of multiplicity of k for the MB, soft and hard data samples (630 and 1800 GeV)
  Probability for finding 1,2,3,4 k in a event. MB, soft and hard data samples are shown.

5. distribution of multiplicity of Lambda for the MB, soft and hard data samples (630 and 1800 GeV)
  Probability for finding 1,2,3 Lambda in a event. MB, soft and hard data samples are shown.

6. invariant pT distribution of k for the MB, soft and hard samples (1800 GeV)
  The inclusive invariant pT dustributions of k (MB, soft and hard) at 1800 GeV.
  We use the form: A( p0 / (pT+p0))exp(n) to fit MB distribution and obtain that
  <pT>=0.74+-0.07 GeV/c

7. invariant pT distribution of k for the MB, soft and hard samples (630 GeV)
  The inclusive invariant pT distributions of k (MB, soft and hard) at 630 GeV.
  With the same form <pT>=0.70+-0.08 GeV/c

8. invariant pT distribution of Lambda for the MB, soft and hard samples (1800 GeV)
  The inclusive invariant pT distributions of Lambda (MB, soft and hard) at 1800 GeV.
  The above form gives <pT>=0.95+-0.09 GeV/c. An exponential function fits the
  data equally well and gives <pT>=1.03+-0.01 GeV/c

9. invariant pT distribution of Lambda for the MB, soft and hard samples (630 GeV)
  The inclusive inavriant pT distributions of Lambda (MB< soft and hard) at 630 GeV.
  The two functions give respectively: <pT>=0.90+-0.07 and <pT>=0.97+-0.01 GeV/c

10. dependence of the mean pT on the event charged multiplicity  (MB, 1800 GeV)
  (the mean pT in figs 10 to 15 is not computed from the fit of the distribution but
  as the mean value of the measured pT of all the k/Lambda observed)
11. dependence of the mean pT on the event charged multiplicity  (Soft, 1800 GeV)
12. dependence of the mean pT on the event charged multiplicity  (Hard, 1800 GeV)
13. dependence of the mean pT on the event charged multiplicity  (MB, 630 GeV)
14. dependence of the mean pT on the event charged multiplicity  (Soft, 630 GeV)
15. dependence of the mean pT on the event charged multiplicity  (Hard, 630 GeV)

16. dependence of the mean number of k on the event charged multiplicity  (1800 GeV)
  Average number of k per event and per charged track, plotted as a function of the
  number of charged tracks (event multiplicity).
17. dependence of the mean number of Lambda on the event charged multiplicity  (1800 GeV)
18. dependence of the mean number of k on the event charged multiplicity  (630 GeV)
19. dependence of the mean number of Lambda on the event charged multiplicity  (630 GeV)
 

Other plots:

1. efficiency(pT): the efficiency for finding K and L0 as a function of their pT
2. efficiency(Lxy): the efficiency for finding K and L0 as a function the secondary vertex
                                displacement in the transverse plain
3. efficiency(t): the efficiency for finding K and L0 as a function of lifetime/<lifetime>
4. efficiency(Nch): the efficiency for finding K and L0 as a function of the charged
                                multiplicity of the event
5. efficiency(eta): the efficiency for finding K and L0 as a function of their pseudo-rapidity