p–Pb collisions

systematic uncertainty at the interpolated energy. For p

_T

>

5 GeV =c, the relative difference between the NLO-scaled spectrum for different choices of the renormalization

_R

and factorization

_F

scales (

_R

¼

_F

¼ p

_T

, p

_T

= 2 , 2 p

_T

) is added to the systematic uncertainties on the spectrum at 7 TeV. In addition, an uncertainty of 2.2% is estimated ny comparing the interpolated and the NLO-scaled data.

The total systematic uncertainty ranges from 7.7% to 8.2%

for 0 : 5 < p

_T

< 20 GeV =c. The NLO-based scaling of the data at ffiffiffi

p s

¼ 2 : 76 TeV gives a result well within these uncertainties. More details can be found in Ref. [16].

The final pp reference spectrum is the product of the interpolated invariant cross section and the average nuclear overlap h T

_pþPb

i , calculated employing the Glauber model [17], which gives h T

_pþPb

i¼h N

_coll

i =

_NN

¼ 0 : 0983 0 : 0035mb

¹

, with h N

_coll

i ¼ 6 : 9 0 : 7 and

_NN

¼ 70 5 mb . The uncertainty is obtained by varying the parameters in the Glauber model calculation; see Ref. [11] (the uncertainties on

_NN

and h N

_coll

i cancel partially in the calculation of h T

_pPb

i ).

The p

_T

spectra of charged particles measured in mini-mum bias (0%–100% centrality, NSD) p þ Pb collisions

at p ffiffiffiffiffiffiffiffi s

_NN

¼ 5 : 02 TeV are shown in Fig. 1 together with the

interpolated pp reference spectrum. At high p

_T

, the p

_T

distributions in p þ Pb collisions are similar to those in pp collisions, as expected in the absence of nuclear effects.

There is an indication of a softening of the p

_T

spectrum when going from central to forward pseudorapidity. This is a small effect, as seen in the ratios of the spectra for forward pseudorapidities to that at j

_c:m:s:

j < 0 : 3 , shown in Fig. 1 (lower panel). We note that several contributions to the systematic uncertainties cancel in the ratios, resulting in systematic uncertainties of 2.2%–5.2% (2.2%–5.9%) for the ratio of the spectrum in 0 : 3 <

_c:m:s:

< 0 : 8 ( 0 : 8 <

_c:m:s:

< 1 : 3 ) to that in j

_c:m:s:

j < 0 : 3 . Calculations with the

DPMJET

event generator [12], which predict well the measured dN

_ch

=d

_lab

[11], overpredict the spectra by up to 22% for p

_T

< 0 : 7 GeV =c and underpredict them by up to 50% for p

_T

> 0 : 7 GeV =c.

In order to quantify nuclear effects in p þ Pb collisions, the p

_T

differential yield relative to the pp reference, the nuclear modification factor, is calculated as

R

_pPb

ð p

_T

Þ ¼ d

²

N

_ch^pPb

=ddp

_T

h T

_pPb

i d

²

^pp_ch

=ddp

_T

; (1)

where N

_ch^pPb

is the charged-particle yield in p þ Pb collisions. The nuclear modification factor is unity for hard processes which are expected to exhibit binary colli-sion scaling. For the region of several tens of GeV, binary collision scaling was experimentally confirmed in Pb þ Pb collisions at the LHC by the recent measurements of observables which are not affected by hot QCD matter, direct photon [18], Z

⁰

[19], and W

[20] production. The present measurement in p þ Pb collisions extends this important experimental verification down to the GeV scale and to hadronic observables.

The measurement of the nuclear modification factor R

_pPb

for charged particles at j

_c:m:s:

j < 0 : 3 is shown in Fig. 2. The uncertainties of the p þ Pb and pp spectra are added in quadrature, separately for the statistical and systematic uncertainties. The total systematic uncer-tainty on the normalization, the quadratic sum of the uncertainty on h T

_pþPb

i , the normalization of the pp data, and the normalization of the p þ Pb data, amounts to 6.0%.

In Fig. 2, we compare the measurement of the nuclear modification factor in p þ Pb to that in central (0%–5%

centrality) and peripheral (70%–80% centrality) Pb þ Pb collisions at p ffiffiffiffiffiffiffiffi s

_NN

¼ 2 : 76 TeV [8]. R

_pPb

is consistent with unity for p

_T

* 2 GeV =c, demonstrating that the strong suppression observed in central Pb þ Pb collisions at the LHC [6–8] is not due to an initial-state effect but rather to a fingerprint of the hot matter created in collisions of heavy ions.

The so-called Cronin effect [21] (see Ref. [22] for a review), namely, a nuclear modification factor above unity at intermediate p

_T

, was observed at lower energies

-2 ) (GeV/c) T dpη)/(dchN2 ) (d T pπ 1/(2evt1/N -7

10 10-6

10-5

10-4

10-3

10-2

10-1

1 10 102

= 5.02 TeV, NSD sNN

ALICE, p-Pb

| < 0.3 ηcms

|

×4) < 0.8 ( ηcms

0.3 <

×16) < 1.3 ( ηcms

0.8 <

| < 0.3 ηcms

pp reference, INEL, |

(GeV/c) pT

1 10

ratio

0.8 1 1.2

| < 0.3 ηcms

< 0.8 / | ηcms

0.3 <

| < 0.3 ηcms

< 1.3 / | ηcms

0.8 <

FIG. 1 (color online). Transverse momentum distributions of charged particles in minimum bias (NSD) p þ Pb collisions for different pseudorapidity ranges (upper panel). The spectra are scaled by the factors indicated. The histogram represents the reference spectrum in inelastic (INEL) pp collisions (see text).

The lower panel shows the ratio of the spectra at forward pseudorapidities to that at j

_c:m:s:

j < 0 : 3 . The vertical bars (boxes) represent the statistical (systematic) errors.

PRL 110, 082302 (2013) P H Y S I C A L R E V I E W L E T T E R S

week ending 22 FEBRUARY 2013

082302-3

Figure 4.10.: Top: Measured differential charged particle yields in non-single-diffractive (NSD) p–Pb collisions in three ranges of ηcms. Results for 0.3 < ηcms < 0.8 and 0.8 < ηcms < 1.3 are scaled for better visibility. Both, statistical and system-atic uncertainties are smaller than the symbol size. The reference spectrum for inelastic (INEL) pp collisions at the equivalent collision energy and identical pseudorapidity range is shown as black line (without uncertainties) in compar-ison. Overall normalization uncertainties of the p–Pb spectra and pp reference are not shown.

Bottom: Ratio of the differential yields measured at moderately forward rapidi-ties to those at mid-rapidity. Statistical uncertainrapidi-ties of the ratio are shown as error-bars and systematic uncertainties as boxes.

Figure published in [164]

4.3. p–Pb collisions 135

To determineR_pPb, the pp reference obtained from the differential cross section is used. 〈T_pPb〉 is calculated from a MC Glauber model with the average taken over all events with at least one binary nucleon-nucleon collision (see section 1.7 for a description of the MC Glauber model). In minimum bias (0-100% central) p–Pb collisions the numerical values obtained with a nucleon-nucleon cross section ofσNN=70±5mb are〈T_pPb〉=0.0983±0.0035mb⁻¹ for the nuclear overlap. The corresponding number of binary collisions is〈N_coll〉=6.9±0.7, andN_part=N_coll+1.

The fact that for p–Pb the spectra are presented in NSD collisions is due to the trigger used in p–Pb (see sections 3.4.1 and 3.8.1 for a description of the trigger). These spectra are compared to inelastic (INEL) pp collisions in terms ofR_pPb. Clearly it would be preferable to to this comparison using also INEL p–Pb collisions. Invariant yields from NSD and INEL p–Pb collisions can differ in shape and normalization.

Differences of the shape are only affecting the lowp_Tpart of the spectrum as diffractive events do not produce high momentum particles in the central rapidity. Estimates with the MC event generator DPMJET show that the difference in shape is below0.5% at the lowestp_T. Above 2 GeV/cin p_Tno difference in shape is observed in the simulation.

The difference in the normalization of the invariant yield to NSD and INEL events, which is the fraction of pure single diffractive events², is significant. With the MC event generators HIJING and DPMJET is was estimated that the INEL yield is 3-4% lower compared to the NSD yield. This number is consistent with the fraction of single-diffractive events measured in pp collisions at 7 TeV (approximately 20%) combined with the fraction of p–Pb collisions in which only a single nucleon-nucleon collision occurs, as estimated from the Glauber MC approach (also approximately 20%). Note that independent of the choice of events in p–Pb collisions (i.e. NSD or INEL), the pp reference should always be for INEL events.

The measuredR_pPb with NSD event selection is shown in Figure 4.11. For p_T>2 GeV/cit is consistent with unity indicating that no strong nuclear effects are present in p–Pb collisions at large transverse momenta. Below 2GeV/cⁱⁿp_Tthe nuclear modification factor drops to about 0.6 at p_T = 0.5GeV/c. A fit to the flat region of the measuredR_pPbin the range6<p_T<20 GeV/cyields an average of

R_pPb

=1.06±0.01(stat.)±0.10(syst.)±0.06(norm.)^.

An enhancement of high p_T particle yields in p–A collisions was first observed in fixed target experiments [199, 200] and is commonly referred to asCronin effect orCronin enhancement.

In d–Au collisions atps_NN= 200 GeV such an enhancement of the nuclear modification factor R_dAu was observed for p_T > 2 GeV/c by the RHIC experiments [201–204]. A maximal of about 1.4 was observed for3 < p_T <5 GeV/c, while lower p_T below 2 GeV/cthe nuclear modification factor was less than unity. For neutral pions, PHENIX measured anR_dAu that is consistent with unity [202].

Only a hint of a Cronin-type enhancement at intermediatep_Tis seen in the minimum bias data presented here. Within the systematic uncertainties the data is also consistent with no Cronin effect at all. A comparison of the results to the nuclear modification factor measured inps_NN

= 0.2 TeV d–Au collisions at RHIC [201, 202] is shown in Figure 4.12.

2 Collisions of p–Pb are called pure single diffractive if they appear like single diffractive nucleon-nucleon col-lisions to the detector. This is the case if all of the underlying nucleon-nucleon collision are single diffractive with all non-diffracting nucleons moving in the same direction.

136 4. Results

(GeV/c) p

T

0 2 4 6 8 10 12 14 16 18 20

pPb

R

0.2 0.4 0.6 0.8 1 1.2 1.4 1.6

1.8 ALICE, charged particles

| < 0.3 ηcms

= 5.02 TeV, NSD, | sNN

p-Pb

Figure 4.11.: The nuclear modification factor of charged particles as a function ofp_Tmeasured in NSD p–Pb collisions atp

s= 5.02 TeV, averaged over the pseudorapidity range

|η_cms|<0.3. Statistical uncertainties are drawn as vertical error bars, systematic uncertainties as open boxed. The filled box at p_T = 0 shows the additional normalization uncertainty of 6%.

(GeV/c) p

T

0 1 2 3 4 5 6 7 8 9 10

dAu

, R

pPb

R

0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2

| < 0.3 ηcms

= 5.02 TeV, NSD, | sNN

ALICE, p-Pb

| < 0.5 = 0.2 TeV, | η sNN

STAR, d-Au

| < 0.18 = 0.2 TeV, | η sNN

PHENIX, d-Au

charged particles

Figure 4.12.: Comparison of the nuclear modification factors measured in p

s_NN = 5.02 TeV p–Pb collisions at the LHC [164] to those measured in p

s_NN = 0.2 TeV d–Au collisions at RHIC by the STAR [201] and PHENIX [202] collaborations. Error bars show statistical and systematic uncertainties added in quadrature. The ad-ditional normalization uncertainties of 6% (ALICE), 17.4% (STAR) and 11.8%

(PHENIX) are shown as filled boxed nearp_T=0.

4.3. p–Pb collisions 137

In Figure 4.13R_pPb is compared to theoretical predictions from various models. Predictions based on gluon saturation models in the Color Glass Condensate (CGC) framework are shown in the top panel of Figure 4.13. The calculation in the framework of the running coupling Balitsky-Kochegov (rcBK) model [205] slightly under-predict the data. The prediction from the Monte Carlo rcBK (rcBK-MC) [206] agrees with the data within the large uncertainties of the model. Calculations from the impact parameter dependent dipole saturation model (IP-Sat) [205] are consistent with the data, as well as other CGC predictions [207] not shown in the figure.

The center panel of Figure 4.13 shows predictions forπ⁰ from next-to-leading order (NLO) pQCD with the EPS09s nuclear parton distribution functions [50] which are in agreement with the data. The EPS09s calculations for π⁰ show no Cronin enhancement, in agreement with measuredR_pPb of identified π^{± [208].} The prediction from leading order (LO) pQCD with cold nuclear matter effects [209] including the isospin effect, Cronin effect, cold nuclear matter energy loss and dynamical shadowing shows a larger suppression at highp_Tthan observed in the data.

In the bottom panel of Figure 4.13 predictions from HIJING 2.1 [210, 211] are shown with different options: default (with shadowing parameters_g =0.28), with decoherent hard scat-tering (DHC), DHC without shadowing, DHC without shadowing and independent fragmenta-tion. None of the HIJING predictions is consistent with the data. With shadowing theR_pPbis under-predicted in a wide range ofp_T and without shadowing a large Cronin enhancement in the intermediatep_Tregion is predicted, that is not seen in the data.

4.3.2 2013 run with extendedp_Trange

With the large statistics p–Pb data sample collected in 2013 thep_Trange of the measurement could be extended up top_T=50 GeV/c. Preliminary results for the measured relative fractions ofπ, K, p were included in the efficiency and acceptance corrections and allowed to reduce the lower threshold to p_T =0.15 GeV/c. Increasing the range also towards low p_T is important for the analysis of〈p_T〉, as this quantity is dominated by the soft part of the spectrum. Hence the results from the 2013 run are used for the〈p_T〉analysis (see chapter 5).

In the analysis of the 2012 pilot run data, the pp reference spectrum for R_pPb covers the same pseudorapidity as the p–Pb measurement (|ηcms| < 0.3). Results from the 2013 run were obtained with the pp reference in a larger pseudorapidity interval (η < 0.8), to mini-mize statistical uncertainties in the reference. The power law parameterization discussed in section 4.2.2 has not been used.

Figure 4.14 shows the R_pPb extracted from the 2013 p–Pb run in the pseudorapidity range

ηcms

< 0.3. The upper plot illustrates the extension to larger p_T. The representation in logarithmic p_T scale (bottom plot) demonstrated the extension of the measurement towards lower p_T and the consistency of the two measurements. At low and intermediate p_T both analysis are in perfect agreement. Small differences are seen in the trend toward large p_T and are partially caused by differences in the reconstruction of simulated collisions used for corrections.

138 4. Results

in proton-nucleus collisions. In ffiffiffiffiffiffiffiffi d þ Au collisions at s

_NN

p ¼ 200 GeV , R

_dAu

reached values of about 1.4 for charged hadrons in the p

_T

range of 3 to 5 GeV =c [23–26].

The present measurement clearly indicates a smaller magnitude of the Cronin effect at the LHC; the data are even consistent with no enhancement within systematic uncertainties.

Data in p þ Pb are important also to provide con-straints to models. For illustration, in Fig. 3, the measure-ment of R

_pPb

at j

_c:m:s:

j < 0 : 3 is compared to theoretical predictions. Note that the measurement is performed for NSD collisions. With the

HIJING

[14] and

DPMJET

[12]

event generators, it is estimated that the inclusion of single-diffractive events would lead to a decrease of R

_pPb

by 3%–4%. Several predictions based on the satura-tion (color glass condensate, CGC) model are available [27–29]. The calculations of Tribedy and Venugopalan [27] are shown for two implementations (running cou-pling Balitsky-Kovchegov (rcBK) and impact parameter dependent dipole saturation (IP-Sat) models; see Ref. [27]

for details). The calculations within IP-Sat are consistent with the data, while those within rcBK slightly under-predict the measurement. The under-prediction of Albacete et al.

[28] for the rcBK Monte Carlo model (rcBK-MC) is consistent with the measurement within the rather large uncertainties of the model. The CGC calculations of

Rezaeian [29], not included in Fig. 3, are consistent with those of Refs. [27,28]. The shadowing calculations of Helenius et al. [30], performed at NLO with the EPS09s parton distribution functions, describe the data well (the calculations are for

⁰

). The predictions by Kang et al. [31], performed within a framework combin-ing leadcombin-ing-order (LO) perturbative QCD (pQCD) and cold nuclear matter effects, show R

_pPb

values below unity for p

_T

* 6 GeV =c, which is not supported by the data.

The prediction from the

HIJING

2.1 model [32] describes, with shadowing, the trend seen in the data, although it seems that, with the present shadowing parameter s

_g

, the model underpredicts the data. The

HIJING

model imple-mentation of decoherent hard collisions (DHCs) has a small influence on the results; the case of independent fragmentation is included for this model and improves agreement with data at intermediate p

_T

. The comparisons

(GeV/c) pT

0 2 4 6 8 10 12 14 16 18 20

PbPb , RpPbR

0.2 0.4 0.6 0.8 1 1.2 1.4 1.6

1.8 ALICE, charged particles

| < 0.3 ηcms

= 5.02 TeV, NSD, | sNN

p+Pb

| < 0.8 η = 2.76 TeV, 0%-5% central, | sNN

Pb+Pb

| < 0.8 η = 2.76 TeV, 70%-80% central, | sNN

Pb+Pb

FIG. 2 (color online). The nuclear modification factor of charged particles as a function of transverse momentum in minimum bias (NSD) pþPb collisions at pffiffiffiffiffiffiffiffis_NN ¼5:02 TeV. The data forj_c:m:s:j<0:3 are compared to measurements [8] in central (0%–5% centrality) and peripheral (70%–80%)PbþPb collisions at pffiffiffiffiffiffiffiffis_NN ¼2:76 TeV. The statistical errors are repre-sented by vertical bars, the systematic errors by (filled) boxes around data points. The relative systematic uncertainties on the normalization are shown as boxes around unity nearp_T ¼0 for pþPb (left box), peripheral PbþPb(middle box), and central PbþPb (right box).

0.4 0.6 0.8 1 1.2 1.4 1.6

1.8 p+Pb s_NN = 5.02 TeV

| < 0.3 ηcms

ALICE, NSD, charged particles, |

Saturation (CGC), rcBK-MC Saturation (CGC), rcBK Saturation (CGC), IP-Sat

pPbR

0.4 0.6 0.8 1 1.2 1.4 1.6

1.8 Shadowing, EPS09s (π⁰) LO pQCD + cold nuclear matter

(GeV/c) pT

0 2 4 6 8 10 12 14 16 18 20

0.4 0.6 0.8 1 1.2 1.4 1.6

1.8 _{HIJING 2.1} ^sg=0.28_=0.28

DHC, sg

DHC, no shadowing DHC, no shadowing and independent fragmentation

FIG. 3 (color online). Transverse momentum dependence of the nuclear modification factor R_pPb of charged particles mea-sured in minimum bias (NSD) pþPb collisions at pffiffiffiffiffiffiffiffis_NN ¼ 5:02 TeV. The ALICE data in j_c:m:s:j<0:3 (symbols) are compared to model calculations (bands or lines, see text for details). The vertical bars (boxes) show the statistical (systematic) errors. The relative systematic uncertainty on the normalization is shown as a box around unity near p_T ¼0.

PRL 110, 082302 (2013) P H Y S I C A L R E V I E W L E T T E R S

week ending

22 FEBRUARY 2013

082302-4

Figure 4.13.:R_pPbmeasured for charged particles in NSD events in comparison to model calcu-lations. Systematic uncertainties of the data are shown as open boxes, statistical uncertainties as vertical bars. The systematic uncertainty of the normalization is shown as a box around unity atp_T = 0.

Figure published in [164].

4.3. p–Pb collisions 139

(GeV/c) pT

0 5 10 15 20 25 30 35 40 45 50

pPbR

0.4 0.6 0.8 1 1.2 1.4

1.6 | < 0.3

ηcms

= 5.02 TeV, NSD, | sNN

ALICE, charged particles, p-Pb, 2012 p-Pb run

2013 p-Pb run

(GeV/c) pT

10-1 1 10

pPbR

0.4 0.6 0.8 1 1.2 1.4

1.6 | < 0.3

ηcms

= 5.02 TeV, NSD, | sNN

ALICE, charged particles, p-Pb, 2012 p-Pb run

2013 p-Pb run

Figure 4.14.:R_pPbfrom the analysis of the 2013 data [173] in comparison to results from the 2012 pilot run [164]. Both measurements are for|ηcms|<0.3, but the pp refer-ence spectrum for 2012R_pPbis taken at

η

<0.3, while the 2013 results use the pp reference in

η

<0.8. Systematic uncertainties are shown as boxes and sta-tistical uncertainties vertical lines. Note that the systematic uncertainties of the two measurements are correlated to a large extend. The top (bottom) plot is in linear (logarithmic) scale inp_Tto demonstrate the extendedp_T reach. The addi-tional normalization uncertainty of 6%, which is common to both measurements and fully correlated between them, is not shown.

140 4. Results

ALI-DER-72

Figure 4.15.: Comparison ofR_pPb from 2013 ALICE data to preliminary results measured by the CMS Collaboration [212]. The pseudorapidity range is slightly different with

ηcms

< 1 for the CMS data and −0.3 < ηcms < 1.3 for the ALICE results.

Overall normalization uncertainties are shown as boxes aroundR_pPb=1. The nuclear modification factor remains consistent with unity also above 20GeV/cⁱⁿp_T. Also the EPS09s calculations [50] remain in agreement with the extended measurement also for p_T>20 GeV/c. Average values ofR_pPbin|ηcms|<0.3in selectedp_Tranges are:

• R_pPb

=0.995±0.007(stat.)±0.084(syst.)±0.060(norm.)^for10<p_T<20 GeV/c

• R_pPb

=0.990±0.031(stat.)±0.090(syst.)±0.060(norm.)^for20<p_T<28 GeV/c

• R_pPb

=0.969±0.056(stat.)±0.090(syst.)±0.060(norm.)^for28<p_T<50 GeV/c Preliminary results onR_pPb at p

s_NN = 5.02 TeV have also been presented by the other LHC experiments CMS [212] and ATLAS [213, 214]. The CMS results are for

ηcms

<1and ATLAS presented R_pPb in the centrality interval 0-90% for the rapidity range

y_cms

< 0.5. Both experiments observe a rise of R_pPb above unity for p_T >30GeV/cand a continuing increase up toR_pPb ≈ 1.4 at p_T =100 GeV/c. This trend is not seen in the ALICE data, but within the statistic and systematic uncertainties of the measurements no discrepancy can be claimed.

A direct comparison of the ALICE results to the CMS measurement is shown in Figure 4.15.

At p_T <20 GeV/c both measurements are in remarkable agreement, towards larger p_T the different trend start to be visible. Differences between ALICE and CMS at largep_Tare present in the p–Pb spectra as well as in the pp reference, but with opposite sign and accumulate in R_pPb.

Im Dokument Transverse momentum distributions of primary charged particles in pp, p–Pb and Pb–Pb collisions measured with ALICE at the LHC (Seite 134-141)