Updated templates for the interpretation of LHC results on

supersymmetry in the context of mSUGRA

arXiv:1202.6580v1 [hep-ph] 29 Feb 2012

K. Matchev, R. Remington1
1
Physics Department, University of Florida, Gainesville, Florida, U.S.A.

(Dated: February 29, 2012)

Abstract

In this short note, we describe the preparation of updated templates for the interpretation
of SUSY results from the LHC in the context of mSUGRA. The standard (m0 , m1/2 ) plane is
shown for fixed µ > 0 and mt = 173.2 GeV. Two scenarios are considered: (1) A0 = 0 GeV
and tan(β) = 10 and (2) A0 = −500 GeV and tan(β) = 40. In each case, the universal scalar
mass parameter m0 varies in the range m0 ∈ [40, 3000] GeV, while the universal gaugino mass
parameter m1/2 varies in the range m1/2 ∈ [100, 1000] GeV. We delineate notable regions in
parameter space, including the region with a charged LSP (stau), the LEP2 reach, and the
cosmologically preferred region with 100% neutralino dark matter. The templates also show
mass contours for a few key particles (gluino, squark and Higgs boson). The mass spectrum is
calculated with the SoftSusy-3.2.4 package, while the neutralino relic density is obtained with
MicrOMEGAs version 2.4.

1
I. INTRODUCTION

Low energy supersymmetry (SUSY) [1–7] is a primary target of the LHC collaborations
in their quest for new physics at the TeV scale. Unfortunately, even the most minimal
supersymmetric extension of the Standard Model comes with a large number of a priori
unknown parameters (superpartner masses, mixing angles and CP violating phases). The
experimental exploration of such a large parameter space is impractical. Therefore, it
has become customary to present experimental limits from searches for supersymmetry in
terms of simple benchmark models with very few input parameters, see e.g. [8–12]. The
most popular SUSY benchmark model is the minimal supergravity (mSUGRA) model [13–
15], a.k.a. the “constrained MSSM” (cMSSM). In this model the SUSY mass parameters
are input at the grand unification (GUT) scale, conventionally defined as the scale where
the two gauge couplings g1 and g2 meet. At the GUT scale, all scalars in the model
have a common mass m0 , while all gaugino fermion superpartners have a common mass
m1/2 . The model has two additional continuous parameters: A0 , a common trilinear scalar
coupling at the GUT scale, and tan(β), the ratio of the two Higgs vacuum expectation
values, plus a discrete parameter, the sign of the higgsino mass parameter µ. In order to
obtain the physical SUSY mass spectrum, the parameters m0 , m1/2 and A0 are evolved via
the renormalization group equations (RGEs) from the GUT scale down to the electroweak
scale, where the radiatively corrected mass spectrum is computed. Since the gauge and
Yukawa couplings of the Standard Model are input at the electroweak scale, the procedure
must be iterated until it converges on a stable solution. There are several state of the art,
publicly available codes on the market which can do these calculations, and they generally
give similar results in the bulk of the parameter space [16]. Here we use the SoftSusy-3.2.4
software package [17].

TABLE I. Relevant input parameters for SoftSusy mass spectrum calculations

Parameter Value
αs (mZ ) 0.1184
−1
αem (mZ ) 127.934
mt 173.2 GeV
mb 4.2 GeV
sign(µ) +
A0 0 GeV -500 GeV
tan(β) 10 40

In this note we provide details of the design of mSUGRA templates which can be
used for interpretation of supersymmetry searches at the LHC. Given the ubiquitousness
of the mSUGRA model, it has become customary to present exclusion limits from SUSY
searches in the m0 −m1/2 parameter plane, for fixed values of the remaining mSUGRA (and
Standard Model) parameters (shown in Table I). In anticipation of the improved reach with
2011 and 2012 data, we extend the previously considered ranges of m0 and m1/2 , allowing
them to vary in m0 ∈ [40, 3000] GeV and m1/2 ∈ [100, 1000] GeV, correspondingly. Two
different m0 − m1/2 slices are considered. The first has a moderate value of tan(β) = 10,

2
 (a) tan(β) = 10 (v1) (b) tan(β) = 10 (v2)

FIG. 1. Template versions 1 (left) and 2 (right) for the tan(β) = 10 scan. The features are
described in the text.

(a) tan(β) = 40 (v1) (b) tan(β) = 40 (v2)

FIG. 2. Template versions 1 (left) and 2 (right) for the tan(β) = 40 scan. The features are
described in the text.

while A0 is taken to be A0 = 0. The second m0 − m1/2 slice has a higher value of

tan(β) = 40 and a different A0 = −500 GeV. In each case, the sign of the µ parameter
is positive, and the top mass is taken to be mt = 173.2 GeV. In what follows, we refer to
these two m0 − m1/2 scans by their respective tan(β) parameters.

II. MSUGRA TEMPLATES

We begin by presenting the templates in their final form before discussing the details
of their respective features. The templates are displayed in Figs. 1 and 2 for tan(β) = 10
and tan(β) = 40, respectively. The calculations are performed across an m0 − m1/2 grid.

3
The LHC mSUGRA scans [18] sample this grid in 20 GeV intervals for both m0 and
m1/2 with about 10k events for each point. A finer granularity is of course desirable
but is computationally prohibitive, as each grid point has to be propagated through the
generation, simulation, and reconstruction chain. The storage space required for hosting
these datasets would also grow quadratically with finer granularity. Since our goal is to
simply provide a canvas which delineates theoretical and experimental regions of interest
in the m0 − m1/2 plane, we only need to make use of the mass spectrum calculations,
and these come in concise SLHA files from the SoftSusy output. This allows us to pursue
a much finer granularity of 5 GeV intervals. In the high tan(β) scan, grid points with
m0 < 220 GeV do not yield a suitable neutral dark matter candidate, thus the scan
effectively begins at m0 = 220 GeV.

III. CONSTRAINTS ON THE MSUGRA PARAMETER SPACE

Within the mSUGRA parameter space considered here, there are regions which are
not considered to be viable for hypothesis testing, since they are subject to constraints
imposed by theory or experiment.

A. Theoretical constraints

1. Charged LSP

One nice feature of the mSUGRA model is that throughout most of the m0 − m1/2 pa-
rameter space, the lightest supersymmetric particle (LSP) is a neutralino, which represents
a potential dark matter candidate. If a suitable discrete symmetry (such as R-parity) is
imposed on the model, the neutralino does become a dark matter candidate and its relic
density can be readily computed from the thermal freeze-out (see Sec. III B 3 below). More
importantly for collider searches though is the fact that neutralinos are very weakly in-
teracting particles, and do not interact inside the detector. Thus mSUGRA parameter
space points with neutralino LSP will be relevant for SUSY searches utilizing the missing
transverse energy in the event.
As can be seen from Figs. 1 and 2, though, at low values of m0 there is a small corner
in the m0 − m1/2 plane (colored dark gray and denoted “τ̃ = LSP ”), where the lightest
particle in the SUSY spectrum is a τ slepton (stau τ̃ ). Being an electrically charged
particle, the stau is not a good dark matter candidate, therefore the dark gray regions in
Figs. 1 and 2 are cosmologically disfavored1 .

2. Correct electroweak symmetry breaking

Another important theoretical requirement on the model is that the global minimum of
the scalar potential should break the electroweak symmetry, leaving QED and QCD intact.
1
This conclusion can be avoided if a) the stau is not the true LSP in the model, but slowly decays
to a superWIMP such as the gravitino [19] or b) if R-parity is violated and the stau decays to SM
particles, with no sizeable missing transverse energy in the event. In either case, the corresponding
collider signatures fall into the scope of the exotica group.
4
Within the m0 −m1/2 plane one does find regions with the wrong gauge symmetry breaking
pattern. For example, at very large values of m0 , one finds an area with no electroweak
symmetry breaking, as indicated by an unphysical value of µ2 < 0. In Figs. 1 and 2
that region is colored light gray and denoted “No EWSB”. (At very small values of m0 at
large tan(β), one may encounter an area with a charge-breaking vacuum, as signalled by a
negative stau mass squared (Mτ̃2 < 0) [20]. For simplicity, we do not specifically delineate
that region, since it is already well inside the stau LSP region.)

3. Convergence on reliable solutions to the RGEs

As already mentioned in the Introduction, the numerical programs used to solve the
MSSM RGEs use an iterative procedure, since some inputs (the gauge and Yukawa cou-
plings and tan(β)) are specified at the weak scale, while others (m0 , m1/2 and A0 ) are
defined at the unification scale. Given that the RGE’s are very nonlinear functions, one
might expect that in some situations the iterative procedure will not converge and instead
will exhibit chaotic behavior. Indeed, this turns out to be the case in a narrow strip
(colored light blue), which is adjacent to the “No EWSB” region in Figs. 1 and 2.
Like all other SUSY spectrum calculators, SoftSusy checks for convergence after each
iteration, and eventually terminates after a given fixed number of iterations2 . One should
keep in mind that even though SoftSusy warns the user of the non-convergence problem,
it still provides a mass spectrum, albeit an unreliable one. This is illustrated in Figs. 3(a)
and 3(b) which show the masses of the two charginos obtained in SoftSusy as a function
of m0 for a horizontal slice across the m0 − m1/2 planes of Figs. 1 and 2 at a fixed value
of m1/2 = 220 GeV. Here, the region of RGE non-convergence as reported by SoftSusy is
shaded in light blue. In this region, the expected value of µ as calculated from minimizing
the radiatively corrected effective potential, is rather small, and the lightest chargino χ̃+1
is higgsino-like. The oscillatory behavior of its mass Mχ̃+1 at high m0 is an indication of
the convergence problem. Since the value of µ in the non-convergence region is small,
this region is always found to buffer the gray “No EWSB” region in which µ2 < 0. It is
important to emphasize that non-convergence of the RGEs by itself does not necessarily
mean that such points are ruled out or disfavored3 — they may very well represent physi-
cally viable models and one should not mistakenly incorporate them into the “No EWSB”
region where µ2 becomes negative. Those are indeed two separate regions and deserve to
be distinguished on the canvas.
In both Figs. 3(a) and 3(b) we observe perfect agreement between the chargino masses
as calculated by our private SoftSusy scan (solid lines, 5 GeV granularity) and those
calculated by the LHC scan [18] (red markers, 20 GeV granularity). While for this slice
of m1/2 = 220 GeV, the LHC scans do not extend into the regions where non-convergent
RGE’s are reported, it does happen in rare cases (a few out of a thousand) — the LHC scan
would sample a point that falls just over the boundary of the region of non-convergence.
We have checked that the chargino masses obtained in those cases do coincide with the
2
We have verified that increasing the maximum number of iterations does not alleviate the convergence
problem, i.e. the chaotic behavior is intrinsically present in the system.
3
Non-convergence of the RGEs means simply that — the mass spectrum calculation is unreliable. In
those cases, the SoftSusy manual [17] recommends using an extrapolation from the neighboring regions
where the program did converge.
5
 (a) tan(β) = 10

(b) tan(β) = 40

FIG. 3. Masses of the two charginos χ̃± ±

1 and χ̃2 (in GeV) as a function of m0 for m1/2 = 220
GeV for the tan(β) = 10 (top) and tan(β) = 40 (bottom) scans. The non-convergent region is
shaded in light blue, while the “No EWSB” region is shaded in gray. The green shaded region is
ruled out by chargino searches at LEP.

result that one would obtain by extrapolating from the convergent region, so no ill-effects
are expected from this occasional trespassing.

It should be noted that while all of these theoretical constraints are ascertained with
the SoftSusy spectrum calculator, they are generally understood and expected constraints
and appear when using other spectrum calculators as well (Suspect [21], SPheno [22],
IsaJet [23], etc.).

6
 B. Experimental Constraints

Results from previous experiments can also be interpreted in the context of the
mSUGRA parameter space shown in Figs. 1 and 2.

1. Chargino mass limits from LEP2

(a) tan(β) = 10 (b) tan(β) = 40

FIG. 4. Lightest chargino mass Mχ̃+ as a function of m0 and m1/2 for (a) tan(β) = 10 and (b)
1
tan(β) = 40.

LEP2 searches for chargino pair production are probing the two areas of the m0 − m1/2
plane which exhibit a light chargino (at low m1/2 , where χ̃+ 1 is wino-like, and at large
m0 where µ becomes small and χ̃+ 1 is higgsino-like). In Figs. 1 and 2 the corresponding
regions are colored green and can be understood as excluded at 95% C.L. by LEP2 [24].
For reference, the lightest chargino mass Mχ̃+1 is plotted across the m0 − m1/2 grid in
Figs. 4(a) and 4(b) for tan(β) = 10 and tan(β) = 40, respectively.

2. Slepton mass limits from LEP2

LEP2 has also performed searches for direct slepton production. Explicitly, the follow-
ing exclusion limits are imposed [24]:
m(ẽR ) < 100 GeV ∧ m(χ̃01 ) < 85 GeV (1)
m(µ̃R ) < 95 GeV ∧ m(µ̃R ) − m(χ̃01 ) > 5 GeV (2)
m(τ̃1 ) < 86 GeV ∧ m(τ̃1 ) − m(χ̃01 ) > 7 GeV (3)
The relatively small region which satisfies this condition can be found at low m0 and low
m1/2 and is colored yellow in Figs. 1 and 2. It can be understood as excluded at 95% C.L.
by LEP2.

7
 3. Dark matter relic density

(a) tan(β) = 10 (b) tan(β) = 40

FIG. 5. Dark matter relic density Ωh2 as a function of m0 and m1/2 for tan(β) = 10 (top) and
tan(β) = 40 (bottom).

In addition to collider constraints from direct searches for superpartners, one may also
wish to consider the cosmological constraint on the dark matter relic density (which is
indirect and perhaps more speculative, since it involves certain assumptions about the
composition of the dark matter as well as the thermal history of the Universe). We calcu-
late the neutralino relic density across the m0 − m1/2 grid with MicrOMEGAs version 2.4
[25] using SoftSusy-3.2.4 for mass spectrum calculations. The resulting Ωχ h2 is displayed
in Figs. 5(a) and 5(b) for tan(β) = 10 and tan(β) = 40, respectively. One can see that
throughout most of the parameter space, the neutralino relic density is quite larger than
the WMAP preferred range [26]
0.102 < Ωχ h2 < 0.123. (4)
Model points satisfying eq. (4) yield the appropriate amount of dark matter to within 3σ
of the measured value of 0.1123 ± 0.0035 and are denoted by magenta filled circles in
version 2 of the templates on Figs. 1(b) and 2(b). As expected, the proper relic density is
obtained in two areas: 1) the neutralino-stau coannihilation region [27] near the boundary
of the “τ̃ LSP” region and 2) the “focus-point” region at large m0 [28, 29], where the
neutralino LSP picks up a non-negligible higgsino component [30]. Other notable features
on the plot in Fig. 5(a) include the opening of the neutralino annihilation channel into
pairs of top quarks (near m1/2 = 430 GeV) and pairs of Higgs and/or gauge bosons (in
the neighborhood of m1/2 = 300 GeV).

IV. MASS SPECTRUM

It is a tradition to display a few meaningful isomass contours for particles of interest

on the mSUGRA canvas. In the past, the gluino and squark masses were displayed in

8
250 GeV intervals.4 We continue to display these contours, opting for 500 GeV intervals
to reduce congestion. In Version (2) of the templates we go further and provide two
additional contours for a Higgs mass of 115 and 120 GeV, respectively. These values are
particularly relevant at this stage as the results of the Higgs searches at the LHC may
soon make contact with this parameter space [31]. As a reference, Figs. 6, 7, and 8 show
the masses of the gluino, squark (average), and light CP-even Higgs boson, respectively,
across the m0 − m1/2 plane.

(a) tan(β) = 10 (b) tan(β) = 40

FIG. 6. Gluino mass in GeV as a function of m0 and m1/2 for (a) tan(β) = 10 and (b) tan(β) = 40.

(a) tan(β) = 10 (b) tan(β) = 40

FIG. 7. Average first generation squark mass in GeV as a function of m0 and m1/2 for (a)
tan(β) = 10 and (b) tan(β) = 40.

4
Here we take the average of the four first generation squarks. These are observed to always be within
8% of one another.

9
 (a) tan(β) = 10 (b) tan(β) = 40

FIG. 8. Higgs mass in GeV as a function of m0 and m1/2 for (a) tan(β) = 10 and (b) tan(β) = 40.

V. APPLICATION

It is expected that Version (1) of the templates is used for public presentation. Both
versions, however, are available as TCanvas objects and C-macros in the ROOT framework.
These can be downloaded from the web at
• tan(β) = 10
tier2.ihepa.ufl.edu/˜remington/SUSY/mSUGRA/GridTanb10 v1.root (.C)
tier2.ihepa.ufl.edu/˜remington/SUSY/mSUGRA/GridTanb10 v2.root (.C)
• tan(β) = 40
tier2.ihepa.ufl.edu/˜remington/SUSY/mSUGRA/GridTanb40 v1.root (.C)
tier2.ihepa.ufl.edu/˜remington/SUSY/mSUGRA/GridTanb40 v2.root (.C)
and are also available as ancillary files from the arXiv version of this note.
The following code can be used in ROOT (v.27) to draw the canvas from the root files
above:

TFile f(‘‘GridTanb10_v1.root’’);
TCanvas *c = (TCanvas*) f.Get(‘‘GridCanvas’’);
c->Draw();
MyExclusionContour->Draw(‘‘SAME’’);

As a final check, we survey the model points that are expected to be in the official
LHC scans [18] and we represent them symbolically with a “+” on top of our templates to
assess the coverage. This can be seen in Figs. 9 and 10 for tan(β) = 10 and tan(β) = 40,
respectively. For tan(β) = 10 we see that there is full coverage everywhere. For tan(β) =
40, however, we see a small envelope in between the dark matter preferred region and the
non-convergent RGE region where the LHC scan does not sample. Our recommendation
is that future scans should be extended to include this interesting region. Analysts should

10
be prepared to perform a small extrapolation in this region in the case that their exclusion
contour terminates nearby.

FIG. 9. LHC scan points (+) overlaying the mSUGRA template for tan(β) = 10.

FIG. 10. LHC scan points (+) overlaying the mSUGRA template for tan(β) = 40.

11
 ACKNOWLEDGMENTS

We thank the CMS SUSY group conveners for encouragement and S. Mrenna and S.
Padhi for useful discussions. We also thank J. L. Feng for pointing out some inconsistencies
in the previous mSUGRA templates used by CMS and ATLAS.

[1] Y. Golfand and E. Likhtman, “Extension of the Algebra of Poincare Group Generators
and Violation of p Invariance,” JETP Lett. 13 (1971) 323–326.
[2] P. Ramond, “Dual Theory for Free Fermions,” Phys.Rev. D3 (1971) 2415–2418.
[3] A. Neveu and J. Schwarz, “Factorizable dual model of pions,” Nucl.Phys. B31 (1971)
86–112.
[4] A. Neveu and J. Schwarz, “Quark Model of Dual Pions,” Phys.Rev. D4 (1971) 1109–1111.
[5] D. Volkov and V. Akulov, “Is the Neutrino a Goldstone Particle?,” Phys.Lett. B46 (1973)
109–110.
[6] J. Wess and B. Zumino, “A Lagrangian Model Invariant Under Supergauge
Transformations,” Phys.Lett. B49 (1974) 52.
[7] J. Wess and B. Zumino, “Supergauge Transformations in Four-Dimensions,” Nucl.Phys.
B70 (1974) 39–50.
[8] M. Battaglia, A. De Roeck, J. R. Ellis, F. Gianotti, K. Matchev, et al., “Proposed
post-LEP benchmarks for supersymmetry,” Eur.Phys.J. C22 (2001) 535–561,
arXiv:hep-ph/0106204 [hep-ph].
[9] B. Allanach, M. Battaglia, G. Blair, M. S. Carena, A. De Roeck, et al., “The Snowmass
points and slopes: Benchmarks for SUSY searches,” Eur.Phys.J. C25 (2002) 113–123,
arXiv:hep-ph/0202233 [hep-ph].
[10] M. Battaglia, A. De Roeck, J. R. Ellis, F. Gianotti, K. Olive, et al., “Updated post -
WMAP benchmarks for supersymmetry,” Eur.Phys.J. C33 (2004) 273–296,
arXiv:hep-ph/0306219 [hep-ph].
[11] S. AbdusSalam, B. Allanach, H. Dreiner, J. Ellis, U. Ellwanger, et al., “Benchmark
Models, Planes, Lines and Points for Future SUSY Searches at the LHC,” Eur.Phys.J.
C71 (2011) 1835, arXiv:1109.3859 [hep-ph].
[12] B. Allanach, M. Bernhardt, H. Dreiner, C. Kom, and P. Richardson, “Mass Spectrum in
R-Parity Violating mSUGRA and Benchmark Points,” Phys.Rev. D75 (2007) 035002,
arXiv:hep-ph/0609263 [hep-ph].
[13] A. H. Chamseddine, R. L. Arnowitt, and P. Nath, “Locally Supersymmetric Grand
Unification,” Phys.Rev.Lett. 49 (1982) 970.
[14] R. Barbieri, S. Ferrara, and C. A. Savoy, “Gauge Models with Spontaneously Broken Local
Supersymmetry,” Phys.Lett. B119 (1982) 343.
[15] L. J. Hall, J. D. Lykken, and S. Weinberg, “Supergravity as the Messenger of
Supersymmetry Breaking,” Phys.Rev. D27 (1983) 2359–2378.
[16] B. Allanach, S. Kraml, and W. Porod, “Theoretical uncertainties in sparticle mass
predictions from computational tools,” JHEP 0303 (2003) 016, arXiv:hep-ph/0302102
[hep-ph].

12
[17] B. C. Allanach, “Softsusy: A program for calculating supersymmetric spectra,” Computer
Physics Communications 143 no. 3, (2002) 305 – 331.
[18] The SLHA files for the official CMS mSUGRA scans can be downloaded from
http://cmssw.cvs.cern.ch/cgi-bin/cmssw.cgi/UserCode/SusyAnalysis/SLHAFILES/mSugraScan/.
[19] J. L. Feng, A. Rajaraman, and F. Takayama, “Superweakly interacting massive particles,”
Phys.Rev.Lett. 91 (2003) 011302, arXiv:hep-ph/0302215 [hep-ph].
[20] J. L. Feng, A. Rajaraman, and B. T. Smith, “Minimal supergravity with m20 < 0,”
Phys.Rev. D74 (2006) 015013, arXiv:hep-ph/0512172 [hep-ph].
[21] A. Djouadi, J.-L. Kneur, and G. Moultaka, “SuSpect: A Fortran code for the
supersymmetric and Higgs particle spectrum in the MSSM,” Comput.Phys.Commun. 176
(2007) 426–455, arXiv:hep-ph/0211331 [hep-ph].
[22] W. Porod, “SPheno, a program for calculating supersymmetric spectra, SUSY particle
decays and SUSY particle production at e+ e- colliders,” Comput.Phys.Commun. 153
(2003) 275–315, arXiv:hep-ph/0301101 [hep-ph].
[23] F. E. Paige, S. D. Protopopescu, H. Baer, and X. Tata, “ISAJET 7.69: A Monte Carlo
event generator for pp, anti-p p, and e+e- reactions,” arXiv:hep-ph/0312045 [hep-ph].
[24] K. Nakamura et al., “Review of particle physics,” Journal of Physics G: Nuclear and
Particle Physics 37 no. 7A, (2010) 075021. http://pdg.lbl.gov.
[25] G. Belanger et al., “Indirect search for dark matter with micrOMEGAs2.4,” Comput.
Phys. Commun. 182 (2011) 842–856, arXiv:1004.1092 [hep-ph].
[26] WMAP Collaboration Collaboration, E. Komatsu et al., “Seven-Year Wilkinson
Microwave Anisotropy Probe (WMAP) Observations: Cosmological Interpretation,”
Astrophys.J.Suppl. 192 (2011) 18, arXiv:1001.4538 [astro-ph.CO].
[27] J. R. Ellis, T. Falk, and K. A. Olive, “Neutralino - Stau coannihilation and the
cosmological upper limit on the mass of the lightest supersymmetric particle,” Phys.Lett.
B444 (1998) 367–372, arXiv:hep-ph/9810360 [hep-ph].
[28] J. L. Feng, K. T. Matchev, and T. Moroi, “Multi - TeV scalars are natural in minimal
supergravity,” Phys.Rev.Lett. 84 (2000) 2322–2325, arXiv:hep-ph/9908309 [hep-ph].
[29] K. L. Chan, U. Chattopadhyay, and P. Nath, “Naturalness, weak scale supersymmetry and
the prospect for the observation of supersymmetry at the Tevatron and at the CERN
LHC,” Phys.Rev. D58 (1998) 096004, arXiv:hep-ph/9710473 [hep-ph].
[30] J. L. Feng, K. T. Matchev, and F. Wilczek, “Neutralino dark matter in focus point
supersymmetry,” Phys.Lett. B482 (2000) 388–399, arXiv:hep-ph/0004043 [hep-ph].
[31] J. L. Feng, K. T. Matchev, and D. Sanford, “Focus Point Supersymmetry Redux,”
arXiv:1112.3021 [hep-ph].

13