1: \documentclass[a4paper,12pt]{elsart}
2: \usepackage{epsfig}
3: \usepackage{array,cite}
4: \input{defs}
5: \def\letter{{Letter}}
6: \bibliographystyle{elsart-num}
7:
8: \begin{document}
9:
10: \begin{frontmatter}
11:
12: \title{
13: Determining the spin of supersymmetric particles at the {LHC}
14: using lepton charge asymmetry.
15: }
16:
17: \author{A.J. Barr}
18: \address{Cavendish Laboratory, University of Cambridge, Madingley Road,
19: Cambridge, CB3 0HE, UK}
20:
21: \ead{alan.barr@cern.ch}
22: \begin{abstract}
23: If signals suggesting supersymmetry (SUSY) are discovered at the LHC then
24: it will be vital to measure the spins of the new particles to
25: demonstrate that they are indeed the predicted super-partners.
26: A method is discussed by which the spins of some of the SUSY
27: particles can be determined.
28: Angular distributions in sparticle decays
29: lead to charge asymmetry in
30: lepton-jet invariant mass distributions.
31: The size of the asymmetry is proportional to the
32: primary production asymmetry between squarks and anti-squarks.
33: Monte Carlo simulations are performed for a particular
34: mSUGRA model point at the LHC.
35: The resultant asymmetry distributions are consistent with
36: a \mbox{spin-0} slepton and a spin-\half\ \ntltwo,
37: but are not consistent with both particles being scalars.
38:
39: \end{abstract}
40: \begin{keyword}
41: Hadronic Colliders, Supersymmetry, Spin, LHC
42: \\Cavendish HEP-2004-14
43: \end{keyword}
44:
45: \end{frontmatter}
46:
47: \section{Spin correlations and charge asymmetry}
48:
49: \EPSFIGURE{spin3.eps, height=4cm}{
50: The decay chain under investigation. The lepton from the \ntltwo\ decay
51: is labeled the `near' lepton, regardless of its charge.
52: Also considered are the diagrams after charge conjugation of the
53: slepton and leptons; or of the squark and quark;
54: or of the entire diagram.
55: \label{fig:feyn}
56: }
57:
58: A recent publication\cite{Richardson:2001df} describes the method by which
59: spin correlations were added to the \herwig\cite{Corcella:2000bw,Moretti:2002eu}
60: Monte Carlo event generator. It includes an example
61: of part of a supersymmetric decay chain,
62: \begin{equation}
63: \label{eq:partchain}
64: \sql \to \ntltwo q_L \to \sslrpm l^\mp q_L
65: \end{equation}
66: in which spin correlations can play a significant role in
67: the kinematics of the emitted particles.
68: When the decay of the slepton is also considered,
69: (\figref{fig:feyn}), the final state consists of two opposite-signed
70: leptons of the same family, a quark jet, and missing energy from the
71: undetected \ntlone.
72: %The Feynman diagram in which the anti-slepton participates
73: %in the decay chain is shown in \figref{fig:feyn}~b.
74:
75: \EPSFIGURE{shapes.eps, height=7cm}{
76: Idealised shapes of the \lqnear\ invariant mass distributions
77: in terms of the re-scaled invariant mass variable
78: $\hat{m}$ defined in eq.~\ref{eq:mhat}.
79: The solid line shows $\frac{dP_\mathrm{PS}}{d\hat{m}}$ (\eqref{eq:Pps}),
80: the dashed line $\frac{dP_1}{d\hat{m}}$ (\eqref{eq:P1}),
81: and the dotted line $\frac{dP_2}{d\hat{m}}$ (\eqref{eq:P2}).
82: \label{fig:shapes}
83: }
84:
85: %Of course this information is not all available to the experimental
86: %physicist, but it is simpler to understand how the
87: %spin correlations effect these distributions.
88: In this \letter\ I examine the observability of spin effects in this decay chain at the LHC.
89: I first consider the distribution of invariant mass of the quark
90: or anti-quark jet from the squark decay,
91: and the lepton or anti-lepton from the \ntltwo\ decay.
92: Using the terminology of \cite{Lester:2001zx},
93: this lepton is referred to as the `near' lepton, as opposed to the lepton
94: from the slepton decay, which is called the `far' lepton.
95: Initially I investigate the parton-level distributions which
96: one would obtain if the sign of the squark and the identity
97: of the lepton were known.
98: In \secref{sec:plusminus} the experimentally observable
99: distributions are introduced.
100: These have added complications due to the difficulty in
101: (a) distinguishing which lepton came from which decay, and
102: (b) measuring the charge of the (s)quark.
103:
104: In the approximation in which the quark and lepton are massless,
105: and the SUSY particles are on-mass-shell, then the \lqnear\ invariant mass
106: has a simple and direct interpretation in terms of
107: the angle, $\theta^*$, between the quark and lepton in the $\ntltwo$
108: rest frame,
109: \begin{equation}
110: \mlqnearsq %%= ( p_l + p_q )^2 = p_l^2 + p_q ^2 + 2 (p_l\cdot p_q)
111: = 2 |{\bf p}_l| |{\bf p}_q|(1-\cos\theta^*)
112: = \mlqnearmaxsq \sin^2(\theta^*/2)
113: \end{equation}
114: where ${\bf p}_l$ and ${\bf p}_q$ are the 3-momenta of the near lepton and the
115: quark respectively in the \ntltwo\ rest frame, and the kinematic maximum of \mlqnear\
116: is given by
117: \begin{equation}
118: \mlqnearmaxsq = (m_\sql^2 - m_\ntltwo^2 )
119: (m_\ntltwo^2 - m_\sslr^2) / m_\ntltwo^2 \ .
120: \end{equation}
121:
122: As one would expect, for events in which $\mlqnear$ is at its maximum value
123: $\sin^2(\theta^*/2) = 1$, \ie\ $\theta^*=\pi$,
124: and the near lepton and the quark are back-to-back
125: in the rest frame of the \ntltwo.
126:
127: If the spin correlations were ignored, and particles
128: were allowed to decay according to phase-space, then
129: the probability density function would be
130: \begin{equation}
131: \frac{dP_\mathrm{PS}}{d\hat{m}} = 2\sin(\theta^*/2) = 2\hat{m}
132: \label{eq:Pps}
133: \end{equation}
134: where the rescaled invariant mass variable, $\hat{m}$, is defined by
135: \begin{equation}
136: \hat{m}\equiv \mlqnear/\mlqnearmax = \sin(\theta^*/2)\ .
137: \label{eq:mhat}
138: \end{equation}
139: Taking spin correlations into account, there are extra spin
140: projection factors in the amplitude
141: of $\sin(\theta^*/2)$ or $\cos(\theta^*/2)$ depending
142: on the helicities of the \anti lepton and \anti quark,
143: so that for
144: $l^+ q$ or $l^-\bar{q}$
145: \begin{equation}
146: \frac{dP_1}{d\hat{m}} = 4 \sin^3(\theta^*/2)
147: = 4 \hat{m}^3
148: \label{eq:P1}
149: \end{equation}
150: while for $l^- q$ or $l^+\bar{q}$
151: \begin{equation}
152: \frac{dP_2}{d\hat{m}} = 4 \sin(\theta^*/2)\cos^2(\theta^*/2)
153: = 4 \hat{m}(1-\hat{m}^2) \ .
154: \label{eq:P2}
155: \end{equation}
156: The functional form of these distributions is shown in \figref{fig:shapes}.
157:
158: \section{Parton level}
159:
160: %% \TABULAR[b]{|l|l|l|l|l|}{
161: %% \hline
162: %% \mnought & \mhalf & $A_0$ & $\tan(\beta)$ & $\mathrm{sign}(\mu)$\\
163: %% \hline
164: %% 100~GeV & 300~GeV & 300~GeV & 2.1 & + \\
165: %% \hline
166: %% }{
167: %% \label{tab:point}
168: %% Definition of the GUT-scale parameters for the point investigated.
169: %% The top quark mass (important in the RGE running) was set to 175~GeV.
170: %% }
171:
172: \begin{table}[b]\begin{center}\begin{tabular}{|l|l|l|l|l|l|l|l|l|l|}
173: \hline
174: \gluino & \ntlone & \ntltwo & \ssul & \ssdl & \sser & \ssel\\
175: \hline
176: 717 & 116 & 213 & 631 & 634 & 153 & 229\\
177: \hline
178: \end{tabular}\caption{
179: \label{tab:mass}
180: Masses of selected particles (in GeV) for the model point investigated.
181: }
182: \end{center}
183: \end{table}
184:
185: %% \EPSFIGURE{sugra_pt5.1200.in.eps, height=10cm, angle=+90}{
186: %% Part of the decay spectrum for the model point invesigated.
187: %% Solid black lines indicate branching ratios (BRs) greater than 10\%,
188: %% dashed blue lines show BRs in the range $1\% \to 10\%$, while
189: %% red dotted lines show BRs in the range $0.1 \to 1\%$.
190: %% The vertical axis shows masses in GeV.
191: %% The sparticles are displaced horizontally for clarity.
192: %% \label{fig:spectrum}
193: %% }
194:
195: I take as an example model the mSUGRA point
196: with $m_0= 100$~GeV, $m_\half=300$~GeV, $A_0=$~300~GeV,
197: $\tan\beta=2.1$ and $\mu>0$.
198: For this point the decay chain (\figref{fig:feyn}) has already been well
199: investigated in the context of sparticle mass
200: measurements\cite{phystdr,Lester:2001zx}.
201: The point was originally motivated by cosmological
202: \ntlone\ relic density considerations,
203: but it has subsequently been excluded
204: by the LEP light Higgs limit\cite{Barate:2003sz}.
205: This has no bearing on the analysis presented in this \letter,
206: since the point is used only to illustrate the method.
207:
208: The SUSY mass spectrum and decay branching ratios were calculated with \isajetv{7.64}.
209: Some of the important sparticle masses are shown in \tabref{tab:mass}.
210: These values differ from those of \cite{phystdr,Lester:2001zx}
211: because a more recent version of the \isajet\ program has been used
212: with updated renormalization group evolution.
213: One significant point to mention in respect to this
214: analysis is that the \ntltwo\ is largely Wino,
215: so the branching ratios $\sqr\to\ntltwo q$ are
216: highly suppressed compared to the equivalent decays for the left-handed squarks.
217:
218: \label{sec:lqnear}
219: To examine the parton-level distributions, a small sample of inclusive SUSY events
220: $(\approx2.5~\invfb)$ was generated using
221: the \herwigv{6.505} Monte Carlo event generator, with the
222: leading-order parton distribution functions
223: of {\tt MRST}\cite{Martin:1998sq} (average of central and higher gluon).
224: There is an obvious difference between the \lqnear\ distributions for leptons
225: and anti-leptons (\figref{fig:near}a).
226: The differences between these shapes and those in the idealized
227: distributions (\figref{fig:shapes}) are largely due to
228: sparticles being off their mass shells and
229: contributions from the various different-mass squarks.
230:
231: The distinctive charge asymmetry is caused by spin
232: correlations carried by the \mbox{spin-\half} \ntltwo,
233: so if one could measure the \mlqnear\ invariant mass distribution
234: in the decay chain (\ref{eq:partchain})
235: it would be easy to show that the \ntltwo\ is \mbox{spin-\half}.
236:
237: \EPSFIGURE{near.eps, height=8.5cm}{
238: Invariant mass distributions of {\bf (a)} \lqnear\ and {\bf (b)} \lqbnear,
239: at the parton level.
240: The triangles are for a negatively charged near lepton,
241: while the circles are for a positively charged near lepton.
242: For the test point the on-shell kinematic maximum is 413.4~GeV.
243: Note that these distributions cannot be measured
244: directly by the experiment.
245: %$\int L dt\approx 2.5~\invfb$.
246: \label{fig:near}
247: }
248:
249: However there are experimental difficulties in making such a measurement.
250: As was noted in \cite{Richardson:2001df}, in the decay of
251: an anti-squark the asymmetry in the lepton charge distributions
252: is in the opposite sense to that from squark decays (see \figref{fig:near}b).
253: As it is experimentally unlikely that one could distinguish between
254: a quark and an anti-quark jet at the LHC,
255: only the sum of the $lq$ and $l\bar{q}$ distributions
256: can be considered to be observable.
257: Thus, if equal numbers of squarks and anti-squarks were produced,
258: the \lqnear\ distribution would be indistinguishable from the
259: phase-space distribution, and no spin information could be obtained.
260: However the fact that the LHC will be a proton-proton collider
261: means that the production of squarks will be enhanced compared to anti-squarks,
262: which can then lead to a significant spin-generated lepton charge asymmetry.
263:
264: \section{$\tilde{q}-\tilde{\bar{q}}$ production asymmetry}
265:
266: In a $pp$ collider, the production processes
267: \begin{equation}
268: q g \to \tilde{q} \tilde{g} \quad \mathrm{and} \quad \bar{q} g \to \tilde{\bar{q}} \tilde{g}
269: \label{eq:pro}
270: \end{equation}
271: will produce more squarks than anti-squarks. This is because the
272: quark parton distribution function (PDF) is larger than
273: that of the anti-quark due to the presence of the valence quarks.
274:
275: For a significant $\tilde{q}-\tilde{\bar{q}}$ asymmetry, the processes
276: (\ref{eq:pro}) must provide a significant
277: contribution to the \anti squark production.
278: That is to say that they must not be much smaller than
279: competing charge-symmetric processes such as
280: \begin{equation}
281: q \bar{q} \to \tilde{q} \tilde{\bar{q}} \quad , \quad
282: g g \to \tilde{q} \tilde{\bar{q}}
283: \end{equation}
284: or gluino pair production followed by decay to \anti squarks.
285: One must also check that the processes (\ref{eq:pro})
286: sample the parton distribution functions
287: in a region of $x$ and $\mu_F^2$ where valence quark PDFs are significant.
288:
289: It is safe to assume that by the time SUSY spin measurements are
290: being made at the LHC, the PDFs will be well measured
291: from other measurements, for example from electroweak boson production.
292: One would expect that the production asymmetry, which is the result of the
293: well measured valence quark distribution,
294: would be insensitive to any remaining uncertainty in the PDFs.
295: %However the study of any remaining PDF dependence is beyond the scope of this \letter.
296:
297: \begin{figure}
298: \begin{center}
299: \begin{minipage}[b]{.49\linewidth}
300: \begin{center}
301: \epsfig{file=mrst98lo.eps, height=6cm}
302: \\ \ \hspace{1cm} {\bf(a)}
303: \end{center}
304: \end{minipage}\hfill
305: \begin{minipage}[b]{.49\linewidth}
306: \begin{center}
307: \epsfig{file=x.eps, height=6cm}
308: \\ \ \hspace{1cm} {\bf(b)}
309: \end{center}
310: \end{minipage}\hfill
311: \caption{
312: {\bf (a)} The parton distribution functions
313: used in this \letter\ plotted at factorization scale $\mu^2_F=1~\TeV^2$ \cite{Martin:1998sq}.
314: {\bf (b)} 2-dimensional histogram of the values of $x_1$ and $x_2$ sampled
315: by the Monte Carlo in sparticle pair production at the mSUGRA point investigated.
316: \label{fig:pdf}
317: }
318: \end{center}
319: \end{figure}
320:
321: The Bjorken $x_{1,2}$ parameters, giving the fraction of the proton momenta
322: carried by the initial state partons satisfy
323: \begin{equation}
324: x_1 x_2 = M^2/s \ ,
325: \end{equation}
326: where $s$ is the usual Mandelstam variable.
327: For the LHC, $\sqrt{s}=14$~TeV,
328: and for the test point the invariant mass, $M$,
329: for squark pair production is about 1.3~TeV,
330: so the $x_i$ can be expected to be of the order of 0.1. This is confirmed
331: from the values of $x$ sampled by the Monte Carlo (\figref{fig:pdf}b).
332:
333: In terms of producing a $\tilde{q}-\tilde{\bar{q}}$ asymmetry,
334: it is advantageous that at $x\approx 0.1$ and $\mu_F^2\sim$~1~TeV$^2$
335: the valence quark parton distribution functions are large (\figref{fig:pdf}a).
336: The over-production of squarks relative to anti-squarks
337: for the test point can be observed by comparing the
338: normalisation of the $lq$ and $l\bar{q}$ plots in \figref{fig:near}, where
339: it can be seen that approximately twice as many squarks are produced
340: as anti-squarks for this point.
341:
342: \section{Experimental observables}
343: \EPSFIGURE{far.eps, height=8.5cm}{
344: Invariant mass distributions of {\bf (a)} \lqfar\ and {\bf (b)} \lqbfar,
345: at the parton level.
346: The circles indicate the distribution for the negatively
347: charged far lepton, while the triangles are for
348: the positively charged far lepton.
349: Note that these distributions, like \figref{fig:near},
350: cannot be directly measured by the experiment.
351: The explanation for the charge asymmetry is given in the text.
352: %$\int L dt\approx 2.5~\invfb$.
353: \label{fig:far}
354: }
355:
356: Although the \lqnear\ invariant mass would provide the theoretically cleanest spin signature,
357: with the experimental data it will not be possible to distinguish the
358: near lepton (from the \ntltwo\ decay) from the
359: far lepton (from the \sslrpm\ decay) on an event-by-event basis.
360: Since all of these signal events will contain two leptons of the
361: same family but of different sign, what one {\em can} do experimentally
362: is to look for asymmetries in the \lqplus\ and \lqminus\ distributions,
363: each of which will contain contributions from both the near and the far
364: lepton.
365:
366: The asymmetry in the \lqnear\ invariant mass was discussed in \secref{sec:lqnear}.
367: One might naively assume that the invariant mass distribution of the
368: far \anti lepton with the \anti quark would be free
369: from spin-correlation effects, since this lepton originates
370: from the decay of a scalar particle,
371: the \sslrpm, which should wash out any spin effects.
372: However the slepton itself has been produced in the decay of the
373: \ntltwo, and so has a boost relative to the quark jet which
374: depends on its charge.
375: To be explicit about the sense of the difference,
376: recall that the near positive lepton favours being back-to-back with
377: a $q$ jet. This decreases the boost of the negative slepton
378: relative to the $q$ jet, and means that on average the invariant mass of the
379: quark with a negative far lepton will be smaller than with a positive lepton.
380: The \lqfar\ invariant mass distributions
381: are shown in \figref{fig:far}a while the equivalent distributions
382: from anti-squark decay, \mlqbfar, are plotted in \figref{fig:far}b.
383:
384: \begin{figure}
385: \begin{center}
386: \begin{minipage}[b]{.49\linewidth}
387: \begin{center}
388: \epsfig{file=plusminus.eps, width=6.6cm}
389: \\ \ \hspace{1cm} {\bf(a)}
390: \end{center}
391: \end{minipage}\hfill
392: \begin{minipage}[b]{.49\linewidth}
393: \begin{center}
394: \epsfig{file=asym.eps, width=6.6cm}
395: \\ \ \hspace{1cm} {\bf(b)}
396: \end{center}
397: \end{minipage}\hfill
398: \caption{
399: {\bf (a)} The $l^+ q$ (squares) and $l^- q$ (triangles)
400: invariant mass distributions, and
401: {\bf (b)} the charge asymmetry $A^{+-}$ (eq. \ref{eq:asym}) at the parton level.
402: These distributions have folded-in the indistinguishability
403: of the near and far leptons, and quark vs. anti-quark jets.
404: %$\int L dt\approx XXX~\invfb$.
405: \label{fig:asym}
406: }
407: \end{center}
408: \end{figure}
409:
410: The \lqplus\ and \lqminus\ distributions contain
411: contributions from both near and far leptons,
412: and from squark and anti-squark decays.
413: The spin information carried by the \ntltwo\
414: causes an obvious difference in shape between
415: these experimentally-accessible distributions (\figref{fig:asym}a).
416: The charge asymmetry in the differential cross-sections
417: is defined here as
418: \begin{equation}
419: A^{+-} \equiv \frac{s^+ - s^-}{s^+ + s^-},
420: \quad \mathrm{where}\quad s^\pm = \frac{d\sigma}{d(m_{l^{\pm}q})}\ .
421: \label{eq:asym}
422: \end{equation}
423: \label{sec:plusminus}
424: This asymmetry, plotted as a function of $m_{lq}$ in \figref{fig:asym}b,
425: is clearly not consistent with zero.
426: This demonstrates that at parton level the
427: asymmetry survives contamination from anti-squark
428: production, and the experimental lack of knowledge
429: of near vs. far lepton.
430:
431: %\TABULAR[b]{|l|l|l|l|}{
432: %\hline
433: %\multicolumn{2}{|c}{unobservable} &
434: %\multicolumn{2}{|c}{observable} \vline\\
435: %\hline
436: %\lqnear & \lqfar & \lqhigh & \lqlow\\
437: %\hline
438: %413.4 & 388.6 & 413.4 & 325.1\\
439: %\hline
440: %}{
441: %Maximum values (in GeV) for the various lepton-quark invariant mass distributions,
442: %using results from \cite{Lester:2001zx}.
443: %I have used the mass of the \ssul\ to represent the \sql\ masses.
444: %\label{tab:edges}
445: %}
446:
447: \begin{figure}
448: \begin{center}
449: \begin{minipage}[b]{.49\linewidth}
450: \begin{center}
451: \epsfig{file=atlplusminusbsugra55.eps, width=6.5cm}
452: \\ \ \hspace{1cm} {\bf(a)}
453: \\ \epsfig{file=atlasymbsugra55_nfiles_6.eps, width=6.5cm}
454: \\ \ \hspace{1cm} {\bf(c)}
455: \end{center}
456: \end{minipage}\hfill
457: \begin{minipage}[b]{.49\linewidth}
458: \begin{center}
459: \epsfig{file=atlasymbsugra55.eps, width=6.5cm}
460: \\ \ \hspace{1cm} {\bf(b)}
461: \\ \epsfig{file=atlmllsugra.eps, width=6.5cm}
462: \\ \ \hspace{1cm} {\bf(d)}
463: \end{center}
464: \end{minipage}\hfill
465: \caption{
466: {\bf (a)} The \lqplus\ (squares) and
467: \lqminus\ (triangles) invariant mass distributions
468: after detector simulation and event selection.
469: {\bf (b)} The solid circles show the lepton charge asymmetry $A^{+-}$ %(\eqref{eq:asym})
470: as a function of $m_{lq}$, again after detector simulation.
471: The shaded rectangles are the parton-level result scaled down by a factor of 0.6.
472: The stars show a cross-check -- the equivalent detector-level asymmetry
473: with spin correlations suppressed.
474: For both of the upper two plots $\int Ldt$~=~500~\invfb.
475: %{\bf (c)} The chi-squared divided by the number (25) of degrees of freedom
476: %as a function of integrated luminosity collected,
477: %for the detector-level asymmetry distribution. The red circles
478: %correspond to the simulation with full spin correlations,
479: %while the blue triangles are for the phase-space cross-check.
480: %The horizontal dotted lines show the confidence with which
481: %the null hypothesis of no asymmetry could be excluded.
482: {\bf (c)} The detector-level charge asymmetry, $A^{+-}$,
483: with spin correlations, using a 150~\invfb\ subset of the data.
484: {\bf (d)} The opposite-sign, same-family dilepton invariant mass
485: distribution after opposite-sign, different-family subtraction.
486: \label{fig:atlasym}
487: }
488: \end{center}
489: \end{figure}
490:
491: To check that this asymmetry is robust, a 500~\invfb\ sample of inclusive SUSY events
492: was generated with \herwig\ at the same mSUGRA point.
493: The events were then passed through the \atlfastv{2.50}\cite{Richter:1998at}
494: detector simulation.
495:
496: The cuts applied are taken directly from a previous analysis\cite{Lester:2001zx}
497: in which it was emphasised that this selection had not been
498: tuned to the particular mSUGRA point under investigation,
499: and so could claim a degree of model-independence.
500:
501: In brief, the event selection requires:
502: \begin{itemize}
503: \item{
504: exactly 2 electrons or muons of the same family with opposite charge, both having $p_T\geq10$~GeV.
505: }\item{
506: four or more jets with $p_T^{j_1} \geq 100$~GeV and
507: $p_T^{j_k}\geq$~50~GeV for $k \in \{2,3,4\}$, where $p_T^{j_i}$ is the transverse momentum
508: of the $i$th jet ordered in $p_T$ such that $p_T^{j_1}>p_T^{j_2}>p_T^{j_3}>p_T^{j_4}$.
509: }\item{
510: missing transverse momentum, $\ptmiss \geq \max(100~\GeV, 0.2~M_\mathrm{effective})$
511: and $M_\mathrm{effective}\geq 400~\GeV$, where
512: $M_\mathrm{effective} = \ptmiss + \Sigma_{i=1}^4 p_T^{j_i}$
513: }\item{
514: dilepton invariant mass, $m_{ll}\leq m_{ll}^{\max} + 1~\GeV$
515: }\item{
516: dilepton plus jet invariant mass, $m_{llq}\leq m_{llq}^{\max}$
517: where $m_{llq} = \min(m_{llj_1}, m_{llj_2})$.
518: }
519: \end{itemize}
520: A more detailed discussion of this selection, including the
521: definition of the kinematic limits $m_{ll}^{\max}$
522: and $m_{llq}^{\max}$ is offered in \cite{Lester:2001zx}.
523: After applying these cuts the remaining Standard Model background
524: has been shown to be much smaller than the SUSY background\cite{phystdr},
525: and so has not been simulated here.
526:
527: For real data, one would expect that charge-blind
528: $m_{l^\pm q}$ and other invariant mass distributions would be investigated first.
529: One would use these to extract information on the mass of the various sparticles.
530: Using this information, the selection parameters could be tuned
531: before making any attempt to measure the lepton charge asymmetry,
532: however no such tuning has been attempted for this \letter.
533:
534: While the shapes of the $m_{l^\pm q}$ distributions (\figref{fig:atlasym}a)
535: are modified by detector simulation and event selection,
536: the cuts are charge-blind, and so the
537: detector-level asymmetry function (\figref{fig:atlasym}b)
538: retains the same approximate shape as at parton level.
539: To help visual comparison the parton-level asymmetry has been
540: scaled down by a factor of 0.6. A decrease in the observed asymmetry
541: can be expected due to pollution from the SUSY background.
542: A more sophisticated analysis could
543: estimate this background as a function of $m_{lq}$ using a
544: same-sign dilepton event selection.
545: Some smearing is apparent, but the fact that the asymmetry
546: is negative at low \mlqnear\ and positive at high \mlqnear\ is very clear.
547: The shape of the asymmetry function therefore strongly favours a spin-\half\ \ntltwo.
548:
549: A nice contrast can be seen by making a cross-check in which
550: the spin correlations have been switched off in the Monte Carlo
551: so that all particles decay according to pure phase-space.
552: This produces no significant asymmetry (\figref{fig:atlasym}b)
553: and reflects in some sense the experimental expectation for a `scalar \ntltwo'.
554:
555: \begin{figure}
556: \begin{center}
557: \caption{
558: }
559: \end{center}
560: \end{figure}
561:
562: Figure \ref{fig:atlasym}b was made with a high-statistics
563: (500~\invfb) sample so that the shape of the distribution could be clearly seen.
564: However such a large sample is not necessarily required to determine that the
565: spin correlations exist.
566: %In \figref{fig:atlasym}c, the goodness of fit to the null hypothesis of no
567: %spin correlations is shown as a function of integrated luminosity.
568: The integrated luminosity required to make the measurement will be different for
569: each SUSY point, depending on the sparticle spectrum.
570: However for the particular mSUGRA point investigated,
571: the spin-generated charge asymmetry can be seen to be inconsistent with zero
572: with 150~\invfb\ of integrated luminosity (\figref{fig:atlasym}c).
573:
574: It is worth adding that further spin information is available
575: from the dilepton invariant mass distribution.
576: This also has a kinematic limit, this time at
577: \begin{equation}
578: (m_{l^+l^-})^2_\mathrm{max} = (m_\ntltwo^2 - m_\sslr^2 )
579: (m_\sslr^2 - m_\ntlone^2) / m_\sslr^2 \ .
580: \end{equation}
581: Since one lepton of either sign is produced,
582: and both are right-handed, one might expect that the
583: angular distribution would show strong spin effects.
584: However the scalar slepton removes all
585: spin correlations between the leptons,
586: and so the $m_{l^+l^-}$ distribution (\figref{fig:atlasym}d)
587: is in very good agreement with the triangular prediction of phase-space (\figref{fig:shapes}).
588: This agreement would be hard to explain except
589: as a result of a heavy scalar particle carrying lepton number,
590: and so would increase confidence in the supersymmetric nature
591: of the particles participating in the decay chain.
592:
593: \section{Conclusions}
594:
595: The method presented shows that it is possible to determine
596: the spins of some of the SUSY partners of Standard Model
597: particles at the LHC.
598: The charge asymmetry lends good supporting evidence
599: to the hypothesis that one is observing supersymmetry
600: and to the identity of the sparticles participating in the decay chain.
601: For the point examined, the shape of the asymmetry
602: distribution could be observed with an integrated luminosity of 150~\invfb.
603:
604: It should be noted that the method presented here requires
605: a particular decay chain (\figref{fig:feyn}) to occur at a reasonably
606: high rate, and an initial asymmetry in the
607: squark vs. anti-squark production cross-sections.
608: However, it could be employed to disentangle supersymmetry from
609: other phenomenologically interesting models,
610: such as universal extra dimensions with
611: Kaluza-Klein parity\cite{Cheng:2002ab}.
612:
613: %stau measurements?
614: %may help in distinguishing near and far edges?
615:
616: \section*{Acknowledgments}
617: I would like to thank my colleagues in the Cambridge SUSY working group
618: for helpful discussions, particularly Bryan Webber and Chris Lester.
619: I am also pleased to thank Robert Thorne, and the members of the Oxford ATLAS particle
620: physics group, especially Claire Gwenlan,
621: for discussions relating to parton distribution functions.
622: I have made use of the physics analysis framework and tools which are
623: the result of ATLAS collaboration-wide efforts.
624: This work was funded by PPARC.
625:
626: \bibliography{thesis}
627:
628: \end{document}
629: