Jacob Siegel Research

sigma about = signal /sqrt(background)

4/26/2016:

In order to get jsut the output of a command use backticks, ie.

`grep Cross-section cross*.txt` 

CMS has collected 3.81 fb^-1 [19] so far for 13 TeV, so we will use that figure.

If we multiply our calculated production cross section we will get the # of dilatons produced in the current 13TeV LHC run. This number will be pure production, in an ideal world. The point of madgraph is to make cuts in an ideal world to see if we can reduce some of our background and see a higher percentage of the produced dilations. 

4/25/2016:

To do a while loop in tcsh the syntax is :

while ( a == b)

stuff

end

The stuff loops forever until you tell it to break using the command break.

The syntax for an if statement is:

if (a == b) then

stuff

endif

Variables are annoying in tcsh and must be taken care of. To set a variable do

setenv Var_name value

if you must refrence this variable later call it by $Var_name to get its value

You are allowed to setenv for a single variable as much as you like, it is the equivilant of var=1;

Math is also annoying in tcsh, to set a variable, lets say x, = to some math expression type

@ x = (a - b)

Note: their must be a space between the math operator and any numbers.

4/21/2016:

We will run all our models on these standards:

alpha s:    0.118

alpha em inverse: 137.035

Note these above are normalized to mass of Z boson

Top Pole mass: 176.7

Bottom Pole Mass: 4.66

Bottom running mass: 4.18

Z Mass: 91.1876

W Boson mass: 80.385

Tau Mass: 1.77682

Top running mass 160.

All of the above in GeV

I am working on getting the madgraph cross section to better duplicate the higgs working group crossection. After updating all the values mentioned above i generate the model for 2 higgs mass values to compare to the CERN working group values, at mh=120, 130. I am generating sigma(pp>h>aa) while CERN only gives sigma(gg>h). At these mass vlaues sigma(pp>h) = sigma(gg>h). We can use the narrow width approximation sigma(pp>h>aa) = sigma(pp>h)width(h>aa). We can get both these values from the CERN working group at 13TeV so we calculate cern's values based on this formula. If I run the online madgraph param_card calculator we get values for width(tot) and Br(h>aa). We can calculate width(h>aa) = Br(h>aa)*width(tot). Using the calculated values we calculate that the width is very different from the cern width. The calculator can be found here: http://madgraph.hep.uiuc.edu/Calculators/sm/sm_calc.html

Using the width calculated in the madgraph calculator madgraph generates a sigma(pp>h>aa) that is about 4000% error at mh=120.

So then I went and used the CERN values for BR(h>aa) and width(tot) insted of the madgraph values. But I found that that the branching ratios in the param_card are purley for looks, changing their value doesn't affect the cross section at all. However changing the total width affects the cross section exactly like you would expect, double width(tot) and sigma(pp>h>aa) halves. 

In playing with heft I found that the mass of the higgs in the heft model files themselves doesn't affect the output, only the param_card files need to be modified. However modifying the couplings in heft does exactly what you expect. Multiply the coupling by to and the cross section increases by 100. ie sigma^2 = coupling.

Attached is an excel file (sheet 2) that contains my results of the madgraph trials. 

4/18/2016:

In order to run madraph and have it output to a text file instead type:

echo "0" | ./bin/generate_events > output.txt

in order to get the cross section type:

grep Cross-section

In order to remove all condor jobs submitted by a single user type condor_rm 

in order to run condor you need to make a file that ends in jdl, make sure to make that you chmod it to make it executable

make the jdl file like this:

universe = vanilla

Executable = /home/siegel/condor_test/test.tcsh

getenv = True

Should_Transfer_Files = NO

Requirements = TARGET.FileSystemDomain == "privnet"

Output = /home/siegel/condor_test/out.txt

Log = /home/siegel/condor_test/log.txt

Queue = 1

WHere the executable will be the file you want to run, in this case i just ran a simple tcsh script. Condor runs the script exactly as if you were to type ./test.tcsh so everything works like normal.

In order to check on the status of your job (and see all other jobs on the computing nodes type condor_q -global)

4/7/2016:

It appears that yesterdays plot was wrong. This is because I was changing the higgs mass wrong. You can't just go into the param_card.dat and change one thing at a time because lots of stuff is interconected, insted you must use one of these online calculators, plug in your values there, it will generate the correct param_card.dat, which you then put in madgraph and use. The heft model says to use this calculator: http://madgraph.hep.uiuc.edu/Calculators/sm/sm_calc.html

4/6/2016:

We generated sigma(pp>h>aa) at a bunch of higgs masses and compared it to the CERN working group at the same higgs masses and got:

So we calculated the ratio between the two and fitted that:

The fit is for the formula:

We fit to the function: y(mH) = e^(A*mH+B)*e^e^(C*mH+B)

A=  0.0015872808417882

B = 0.663055987267111

C= -0.0537115347386501

D= 1.14046723708183

Using this fit, and propagation of errors we get a P-Value: 0.480794998,

r^2 = 0.946

This assumes the errors from the cern working group are the pdf + as, which we arn't sure about.

More info can be found in our power point:

https://docs.google.com/presentation/d/1hvVu-IELhMDZZTLNSxcIhV1hlrNuCzpYTR9H__G2SQk/edit?usp=sharing

4/4/2016:

I'm working on writing a script that will go and modify the cards and everything I need for madgraph in part to prepare for when we need to run for a bunch of points in the mx,f plane and so i can run over many different higgs masses to characterize their differences from the cern cross sections. If we have a text file test.txt that for example reads:

-1 mijj

24 aa

32 rp

and we just want to change the 32 rp line to 10 rp we use the following command:

sed -r 's/[0-9]+ rp/10 rp/' <test.txt > test1.txt

this will leave the old file untouched and make the change and put it in a new file called test1.txt. It is worth noting that if you make the two files the same, when you try to open the file later it will just stall indefinatly, so moral of the story is to make the two different. If you really need to modify a file but have the same name type:

sed -r 's/[0-9]+ rp/10 rp/' <test.txt > test1.txt

rm test.txt

mv test1.txt test.txt

refrence [18] will let us check production cross sections for various higgs masses.

In order to just replace a line number N type:

sed 'N s/.*/replacement text/' <file1 > fil2

in order to run ./bin/generate_events without having to type the 2 zeros just do:

echo "0" | ./bin/generate_events

From [1] the greatest value of mx we can test for is 150 GeV because our approximation of x>aa as being the only decay breaks down.

4/3/2016:

We encountered several issues with trying to verify madgraph works. We first ran the tutorial (http://www.niu.edu/spmartin/madgraph/madhiggs.html) for higgs with background and duplicated their cross sections for 13 tev. Then we wanted to make sure that we could reproduce the cern figures for 13 tev found here: https://twiki.cern.ch/twiki/bin/view/LHCPhysics/CERNYellowReportPageAt13TeV [17], so we ran madgraph for  just p p > h at a 125.09 GeV higgs and got a cross section of 19.03 pb, compared with the CERN one of 44 pb we have a huge error. So then we realized that in the run_card.dat madgraph is making several generator level cuts that cern probably isn't making, so we change all cuts to -1, which means no cut at all and re run the code and get a new cross section of about 21 ish GeV so the cuts are decreasing our cross section but not enough to account for the difference. We tried changing the masses of the particles in the model to their pole masses instead of the default, but it didn't seem to make a difference so we just leave them at default from here on out.

4/1/2016:

Chris says that when we calculate the cross sections we need to change the sigma(pp>h) and all the other parameters that depend on higgs cross sections and higgs decay widths so that the higgs mass is equal to the dilaton mass we are testing at, which means that we need to change the higgs mass in the formulas and the higgs mass in the heft model.

Forget about VBF, and use this formula:

so for this formula we need to take BR(hSM > aa) at what ever higgs mass = the dilaton mass we run the model at.

Jet matching should only be an issue for pithia, so run without jet matching and run with it and see if the pythia difference is great enough, if so, run for all values on m,x plane, if not a problem forget about it.

We also know that since we are doing madgraph only for leading order, generate madgaph cross sections for sigma(pp>h>aa) (for higgs NOT dilaton) and compare to the sigma on the higgs working group. Generate this ratio k=sigma(actual from higgs working group)/sigma(from madgraph), and rescale the dilaton cross section we generate from our dilaton (deft) model by k so we are confident that we account for NLO. We need to generate this k for all the diffrent descrete values of cross section given by the higgs working group, and if we have time fit a function to k and use that to generate our own descrete parameter space.

3/28/2016:

A useful search engine for looking up high energy physics (HEP) papers: https://inspirehep.net/

In madgraph when we do:

generate    p p  > h > a a 

add process p p > h > a a j j 

we encpounter issues with double counting, called jet matching. This issue was brought up by [11].

To solve this issue we will need to go through [12] and [13]

Coupplings given by [14]

To test and see if jet double counting was the reason our cross sections were off yesterday I just run the heft model wihout vbf (pp>h>aa) and get a cross section of 0.02625 pb, witch when multiplied by our ratio is 0.034 pb

and just the deft (dilaton) model for (pp>h>aa) and get a cross section of: 0.007116 pb, so clearly our model is still very wrong, because we we expect some error due to our neglection of VBF but this error is enormous, and way to big to be from neglecting VBF which is only about a few percent of the total cross section.

3/27/2016:

It appears that the way we can reproduce the exclusion plot is to use the detection ratio outlined in [1].

The cross section returned for madgraph for p p > h > a a j j (VBF) is 0.00831 pb, so basically it means that our cross sections with VBF and without it should be almost identical, which is the case. H > a a with VBF is 0.025 pb

and without it is 0.02626 pb. So the difference is consistant with a VBF cross section of 0.00831 pb.

To test our model we test at f=1TeV and mx = 100GeV:

If i multiply the cross section returned for madgraph for p p > h > a a j j (VBF) which is 0.02501 pb by the ratio (sigma(pp>h)gamma(h>aa)/(sigma(pp>x)gamma(x>aa)))^-1 we get the value of 0.03248 pb. 

I multiply the sigmma(pp>h) in madgraph by sigma(pp>x)/sigma(pp>h) = 8.25 and i multiply the gamma(h>aa) by gamma(x>aa)/gamma(h>aa) = 0.15, after i make these modifications and run the heft model again for VBF I get a cross section of: 0.0591 pb

3/26/2016:

In our meeting Chris outlined that we should use the unmodified heft model to generate a cross section, and multiply by this formula:

in order to obtain the dilaton cross section, then modify the coupplings for the higgs in the heft model so the heft model is effectivly a dilaton model then, generate the cross section. If these 2 are the same then we are confident that out modifications to the model are accurate. 

The formulas for cross sections we have include VBF (vector boson fusion) so in order to generate VBF in madgraph type:

generate p p > h > a a ;this is the normal method of creating a higgs

add process p p > h > a a j j ;this adds tones more fyneman diagrams because jet is such a large class, for example it adds the following:

I ran this higgs model with VBF and get the cross section of 0.02501 pb. 

The plot for invatient mass for pure physics looks like:

Now we still need to modify the cross sections. In our previous logbook we produced a 3D graph that will allow us to calculate the production cross section at any value of m,f. We still need to find the decay cross section.

3/25/2016:

Meeting with Chris:

For the Rg loop, loop over only colored particles, not whatever was stated previously..

for the Rgamma loop loop over only electrically charged particles.

Now we get the same graphs as Rg for the paper: (the reason that we were wrong before was that we were treating the W and tau as quarks, and we were neglecting other quarks.)

From the paper:

From our own mathmatica model:

The formula used to generate this Rg graph is:

If we now make a 3D graph with the cross section on the vertical vs. f (0,3000) and dilaton mass (0,1000):

The mathmatica code can be found on this github. https://github.com/jdavis36/Dilaton

3/23/2016:

To calculate Rg use these formulas:

3/21/2016:

From [7] v = 246 GeV

higgs cross sections for various higgs masses can be found here: https://twiki.cern.ch/twiki/bin/view/LHCPhysics/CERNYellowReportPageAt7TeV

from [7] sigma(gg>x) = Rg(v^2/f^2)

sigma (x>yy) = v^2/f^2 sigma (h>gg) where mass of higgs = mass of dilaton

to calculate Rg neeed for sigma(gg>x) where x = diliton

look at [1] bottom of page 4, (5) and (6), and [7] (6).

for [7] (6) we sum over the t and W.

3/10/2016:

This is a condensed version of the tutorial: http://www.niu.edu/spmartin/madgraph/madhiggs.html (which is far more complete) to make Jeff happy.

To make the graphs I made yesterday after you have downloaded delphes type (not from now on ; represents a comment):

./bin/mg5_aMC ;to run madgraph

import model heft ;this imports the heft model so madgraph can simulate higgs

generate p p > h, h > a a ;this generates higgs using the heft approximation

output higgs_data ;this outputs the data to the higgs_data folder

quit ;to quit madgraph

cd higgs_data ; go into higgs data folder

cd Cards

emacs me5_configuration.txt ;

nb_core =1 ;change from # nb_core = None so we don't use up all available cores in the server

emacs plot_card.dat ;to modify plots made

a 0 # Class number 3 ;add this under the class #2

;between the lines begin and end plot defs delete everything and replace it with this:

pt 1 1

pt 3 2

mij 3 2

;leave all defaults but modify the lines betweenBegin PlotRange and End PlotRange to say:

pt 5 0 200

mij 2 0 200

; then cd back to the higgs_data folder

./bin/generate_events

1

3

0

0 ;0 as many times as you need

; firefox will open and you can see pure physics under madgraph and a general cylindrical detectors simulator under delphes

to do the same analysis for the higgs background:

./bin/mg5_aMC

generate p p > a a / h ;this generates all p p to a a that dont include the higgs 

output higgs_background

cd higgs_background

;modify the cards like last time

./bin/generate_evetns

1

3

0 

0 ;0 as many times as you need

; firefox will open and you can see pure physics under madgraph and a general cylindrical detectors simulator under delphes

To plot both higgs and background on the same plot 

./bin/mg5_aMC

generate p p > a a / h ;this generates all p p to a a that dont include the higgs 

add process p p > a a / h ;this generates all p p to a a that dont include the higgs 

output higgs__data_and_background

cd higgs_data_and_background

cd Cards

emacs run_card.dat

;change number of events from 10000 to 100000

; change the lines (line 114) so they read the following: 35 = pta ! minimum pt for the photons

; and lines (line 175) 90 = mmaa ! min invariant mass of gamma gamma pair

emacs plot_card.dat

a 0 # Class number 3 ;add this under the class #2

mij 3 2 ;delete all plots except this line

mij 2 100 200 ;modify the plot range line so it reads this

./bin/generate_evetns

1

3

2

0 

0 ;0 as many times as you need

; firefox will open and you can see pure physics under madgraph and a general cylindrical detectors simulator under delphes

We modify the higgs couppling to 2 gluons artifically by a factor of 2 so we can more easily see the peak.

cd models

cd heft

emacs couplings.py

change line 13 of the file from

value = '-(AH*complex(0,1))',

to

value = '-2.0*(AH*complex(0,1))

Then run the analysis exactly the same as for higgs_with_background

and you will get the following graphs for pure physics:

and for delphes you get:

To put the higgs and the background at the same time, type (this is for 100000 events)

to make 100000 events edit the run_card.dat:

• generate p p > h, h > a a 

• add process p p > a a / h 

These are the graphs for pure madgraph physics, which means that there is no detector simulator.

For the Delphes detector simulator the graphs look like this. 

The delphes plots for today are all for a general cylindrical detector.

If you are in your run's directory and type root, then in root type TBrowser b, open up the run, then events, then tag_1_delphes_events, you can open root trees and leafs for the data.

To run pythia, pgs, and delphes when prompted to enter type 1 2 3 0.

If unclear please go to tutorial: http://www.niu.edu/spmartin/madgraph/madhiggs.html

3/8/2016:

Use super lhc lumonisity to make dilatons.

3/7/2016:

The gg>x>yy channel looks the good to study in order to exclude dilaton's on the (m,f) plane (m = mass dilaton, f = scale of symmetry breaking, analogs to sm higgs VEV), this is suggested by paper [7] because the best channels to study are "γγ, WW → lνlν, and ZZ → 4l, llνν" 

gg being the most important because the ratio of branting ratios of dilaton to higgs is > the ratio of branching ratios for higgs to WW (which is also equal for ZZ). This paper [7] suggests that for mass of dilaton below 134 GeV this channel is the only channel that may need to be taken into consideration. Figure 6 in the paper shows the exclusion zone for just the yy channel.

In order to generate our signal in madgraph we need heft, so that we can simulate p p > h creation, madgraph can not do this default because of a toploop. To use heft in madgraph type import model heft. Before heft we got an error with p p > h but not it runs smooth. You must do this everytime you open madgraph, because models are not loaded by default. When I run for p p > h, h > a a (note: a is a photon in madgraph). I get a cross section of 0.02626+-2.3e-5 pb

and pt graphs that look like this:

The peaks of these 2 plots are below the higgs mass, but since the mass of the higgs is split between the 2 photons, the values of the 2 peaks added are 125 GeV.

Notes on using the heft model:

http://www.niu.edu/spmartin/madgraph/madhiggs.html

This has instructions on how to make higgs plots, how to make higgs backgrounds and how to modify higgs coupplings.

Going through the tutorial:

generate p p > h, h> a a

3/4/2016:

We are back working on the dilaton, we need to 1. Find channels that look good

2. Simulate SM background

3. Simulate signal (m,f)

Our goal is to recreate the exclusion graph already produced in papers.

3/2/2016:

To download MadAnalysis run madgraph by ./bin/mg5_aMC

then type: install MadEvent

In order to plot the results of the madgraph simulations, we need to download pythia-pgs. 

run madgraph by ./bin/mg5_aMC

and then: install pythia-pgs

cd into your run's directory, for the tutorial this will be MY_FIRST_MG5_RUN

now run the madevent: ./bin/madevent

now in madevent type: pythia MY_FIRST_MG5_RUN

Now in the firefox browser that opens up look under the  MY_FIRST_MG5_RUN section and under output you should see plots. In this link are all the graphs. 

3/1/2016:

The command quit will quit the madevent/madgraph.

2/29/2016:

We are no longer working on the dilaton search, but are insted making monte-carlo's. To do this we use a tool called madgraph. To obtain madgraph register for an account here: http://madgraph.hep.uiuc.edu/

under downloads. Download the file, and upload the tar file to the cluster. Once in the same directory as the tar file type tar -zxvf MG5_aMC_v2.3.3.tar.gz

You will get hit with a wall of text. This will have just un-tarred madgraph. To run it type ./bin/mg5_aMC

once madgraph opens type tutorial to go through the tutorial. 

Going through the tutorial:

generate p p > t t~ generates a new process. When I run mine i get 5 processes with 7 diagrams.

Some helpful text from the tutorial:  You can find more information on supported syntax by using:

        MG5_aMC>help generate

        To list all defined processes, type

        MG5_aMC>display processes

        If you want to know more about particles and multiparticles present,

        write

        MG5_aMC>display particles

        MG5_aMC>display multiparticles

in the tutorial I added a process and then try to export my process for MadEvent with the command output MY_FIRST_MG5_RUN but when I enter the command I get the following error: 

Error detected in "output MY_FIRST_MG5_RUN_1"

write debug file MG5_debug

If you need help with this issue please contact us on https://answers.launchpad.net/madgraph5

MadGraph5Error : No Fortran Compiler detected! Please install one

This madgraph still creates the folder, so i go intot MY_FIRST_MG5_RUN and emacs me5_configuration.txt and change nb_cores = None to nb_cores = 1

This makes it so that my process is run on only one core at a time and not all, taking up everybody's resources.

To resolve this error make sure to untar the MG5_aMC in the CMSSW_5_3_30 directory you've made, and make sure that before you do anything, as soon as you cd into the MG5 directory to type the cmsenv command. Still follow the rest of the instructions as normal. After changing the code in the Cards directory cd back to the MY_FIRST_MG5_RUN directory and type ./bin/generate_events

This should either open up a firefox browser with the cross section for your run, or if it prompts you with a bunch of options just hit enter until the firefox browser opens. It should look something like this. 

output 

2/16/2016:

I moded the privalages on a few other random files downloaded from the CMS github and all of them return some sort of error with (. 

Located a paper excluding values for nambu-goldstone bosons at tevatron http://arxiv.org/pdf/0909.1319v2.pdf

Exclusions for LHC-LEP combined data:

http://arxiv.org/pdf/1111.3276v2.pdf

This paper seems optimistic that we could find a low mass dilaton with LHC data. 

2/14/2016:

Related CERN twiki:

https://twiki.cern.ch/twiki/bin/view/CMS/SWGuideCMSDataAnalysisSchool2014HggExercise

https://twiki.cern.ch/twiki/bin/view/CMS/Higgs2G

https://twiki.cern.ch/twiki/bin/view/CMSPublic/WorkBook

CMS Github for H->yy:

https://github.com/cms-sw/cmssw/tree/CMSSW_8_0_X/HiggsAnalysis/HiggsToGammaGamma

To get the cms analysis from the CMS github:

mkdir cmsdas2014hgg

cd cmsdas2014hgg

cmsrel CMSSW_6_2_12

cd CMSSW_6_2_12

git clone https://github.com/cms-sw/cmssw.git

cd cmssw

cd HiggsAnalysis

cd HiggsToGammaGamma

Today we copied the CMS github to our cluster accounts, we also found the twiki for the H->yy analysis. 

We tried to follow the instructions in the twiki but we could not locate the /afs/cern.ch/work/b/bendavid/cmspublic/CMSDas2014HGGLongExerciseUpdated/

folder where the root files for the analysis are located, so we got stuck. We then went up to the preexcercises and started from the very beginning of the tutorials. The very first thing the tutorials requite you to do is copy a program from afs: cp /afs/cern.ch/cms/Tutorials/TWIKI_DATA/runThisCommand.py .Which we couldn't do because we couldn't find afs. The tutorial was: https://twiki.cern.ch/twiki/bin/view/CMS/SWGuideCMSDataAnalysisSchoolPreExerciseFirstSet

So we then found the CMS githib, and copied all of that over to our accounts. In the github there was a HiggsToGammaGamma folder so we opened that up and it contained a file called PhotonFix.cc. Even after modifying executable privileges (chmod u+x PhotonFix.cc) when attempting to run the file we still get the error: 

//: Permission denied.

Badly placed ()'s.

This is where we got stuck.

2/12/2016:

[5] The diphoton rate from the 125GeV particle observed at CERN is 2x what they expect from higgs. The technicolor (Note: THIS REFRENCE IS REFERING TO A TECHNICOLOR DILATON, NOT A NAMBU-GOLDSTONE) dilaton couppling is related to the SM higgs couppling by (dilaton ww/zz)/(SM Higgs ww/zz) = vew/Fφ where vew approx = 246GeV, and Fφ =dilaton decay constant. (dilaton ff)/(SM Higgs ff) = vew/Fφ is also true for f = Tau, top, bottom. Fφ^2Mφ^2 = −16Evac, where Evac is the vacuum energy density. This implies that these ratios become lower linearly with the mass of the dilaton, so the lower the mass the harder it will be to distinguish from the higgs. "TD couplings to W W , Z Z and f ¯ f are substantially smaller than those of the SM Higgs. ... The gluon fusion (GF) production at the LHC thus becomes larger than the SM Higgs case due to this extra factor"

We need to establish a lower limit on the dilaton mass. The dilation behaves like a higgs (exactly in couplings but not in overall strength), and Professor Chacko believes nobody has established a lower limit for the gamma gamma channel. This hasn't been done for the LHC below 100GeV because we had already established a lower limit for the Higgs above 100GeV at previous accelerators (Tevatron). We need to access the full data set for the LHC run, find out how to do the higgs gamma gamma analysis. We need to see how low we can set the mass of the higgs below 100 and still get decent background rejection. We will need to do new monte carlo's for below 100GeV as the backgrounds are different and become larger (apparently). 

We are not working with the string theory dilaton but the Nambu-Goldstone dilaton. It is a boson. 

http://arxiv.org/pdf/1111.2580v2.pdf

https://books.google.com/books?id=Lk-7CgAAQBAJ&pg=PA339&lpg=PA339&dq=nambu+dilaton&source=bl&ots=fw59lxPGvd&sig=vSMYXgdQIC94wO1AcZxQBBO0jGc&hl=en&sa=X&ved=0ahUKEwj14YStrO_KAhWBpR4KHQlBDkUQ6AEIMzAC#v=onepage&q=nambu%20dilaton&f=false

My summary of the http://arxiv.org/pdf/1111.2580v2.pdf paper:The Dilaton is a scalar particle appearing in spontaneous scale symmetry breaking at a TeV scale, mostly technicolor theories. It is a boson with spin 0. The Dilaton may be found in Higgs Boson searches because it couples to SM fermions and weak boson exactly the same as the Higgs. This means the dilaton searches can use the existing higgs analysis. It can also couple to a Higgs pair. The mass of the dilaton will be approximately equal to the scale of weak symmetry breaking which will be much less than the scale of scale symmetry breaking (TeV scale). The coupling of the Dilaton to SM particles is the same except that the Standard Model vacuum expectation value (SM VEV) for all parameters is replaced by the TeV scale of scale symmetry breaking. The SM VEV is what you would expect standard model values to be in a pure vacuum, for the higgs this would be 246GeV. The main decay channel of interest is the dilaton to gamma gamma channel. The easiest way to differentiate between the higgs and dilaton is the difference between gluon gluon and gamma gamma couplings, they are enhanced with comparison to higgs. We can also compare the gamma gamma to WW decay channels, their ratios are off compared to the Higgs because their coupling is equal for gamma gamma unequal for WW. Another way to distinguish higgs from dilaton is dilaton can decay to two higgs to 4Z. We are so interested in the higgs like aspects of the dilaton because we want to look for a higgs at below 100GeV and if we see anything it will be a dilaton. Even though the Higgs has a mass of about 125GeV it is still possible via QM that it can exist at these low (sub 100GeV) energies so we must know how to distinguish a dilaton from the Higgs.

               The scale of scale symmetry breaking < 1TeV is unfavorable but there are still regions where this is possible. The 2 free parameters used in the Dilaton theory is the scale of symmetry breaking, and the mass of the dilaton. Meaning that we can only exclude combinations of the symmetry breaking value and the mass together. If plotted vs. each other it is easy to see that values for the symmetry breaking value <1TeV are ruled out at a 95% confidence level for almost every dilaton mass using ATLAS data.

               The motivation for looking for a dilaton is that it is a result of scale symmetry breaking. This symmetry breaking can be traced back to a QCD anomaly, which is resolved by adding a dilaton field. This scale symmetry breaking is tied to chiral symmetry breaking (meaning something does not behave the same for spin up and spin down particles). We are interested in relating the QCD anomaly to the chirality symmetry breaking, which is done via the Dilaton.

Paper [4] says that for a very generous estimate of f (the value of the symmetry breaking) of 1GeV, we expect from pp collisions at the LHC 1.9 dilatons per year. This number is so low that even if we work with many years (generous estimate 5) that is only 10 dilatons. Due to detector inefficiencies and background it seems unlikely that we see any dilatons. For a more conservative estimate of f (5TeV) we have only 0.07 of a dilaton produced per year, and we will definitely not see any. This assumes a dilaton mass of 125GeV.

Also from [4]  Our results demonstrated that for a light dilaton the signal-to-background ratio will be very small, which implies that the probe of a light dilaton in exclusive processes at LHC will be not possible. In contrast, our results indicated that for a massive dilaton, which can be produced if f > 4.0 TeV, the signal-to-background ratio will be favourable if the dilaton decay channels into ZZ and/or HH are considered. Our main conclusion is that the study of exclusive processes can be useful to search a massive dilaton, as well as to constrain the mass and the conformal energy scale.

References:

[1]    http://arxiv.org/pdf/1111.2580v2.pdf        Dilaton at the LHC

[2]    https://books.google.com/books?id=Lk-7CgAAQBAJ&pg=PA339&lpg=PA339&dq=nambu+dilaton&source=bl&ots=fw59lxPGvd&sig=vSMYXgdQIC94wO1AcZxQBBO0jGc&hl=en&sa=X&ved=0ahUKEwj14YStrO_KAhWBpR4KHQlBDkUQ6AEIMzAC#v=onepage&q=nambu%20dilaton&f=false        Dilatons in Dense Baryonic Matter

[3]    http://arxiv.org/pdf/1501.06354v1.pdf        Probing the dilaton in central exclusive processes at LHC

[4]    http://arxiv.org/pdf/1410.2225v1.pdf         Constraining the Higgs-Dilaton with LHC and Dark Matter Searches

[5]    http://www.sciencedirect.com/science/article/pii/S0370269313000567               Is 125 GeV techni-dilaton found at LHC?

[6]    http://arxiv.org/pdf/1209.2017v2.pdf        Holographic techni-dilaton at 125 GeV (Note: THIS REFRENCE IS REFERING TO A TECHNICOLOR DILATON, NOT A NAMBU-GOLDSTONE) 

[7] http://arxiv.org/pdf/1111.3276v2.pdf Exclusions from LHC, LEP combined

[8] http://arxiv.org/pdf/1002.1721.pdf Phenomonology of a light scalar: the dilaton

[9] http://arxiv.org/pdf/1401.5586.pdf higgs dilaton system in LHC data

[10] http://arxiv.org/pdf/1209.3259.pdf ressonance at 125 gev higgs or dilaton

[11] http://www.niu.edu/spmartin/madgraph/madsyntax.html Madsyntax tutorial

[12] https://cp3.irmp.ucl.ac.be/projects/madgraph/wiki/IntroMatching Intro Jet matching

[13] https://cp3.irmp.ucl.ac.be/projects/madgraph/wiki/Matching jet matching

[14] https://cp3.irmp.ucl.ac.be/projects/madgraph/wiki/Models/HiggsEffective heft coupplings

[15] http://arxiv.org/pdf/0706.2334.pdf Some info on heft

[16] https://twiki.cern.ch/twiki/bin/view/LHCPhysics/CERNYellowReportPageAt7TeVhttps://twiki.cern.ch/twiki/bin/view/LHCPhysics/CERNYellowReportPageAt7TeV higgs coupplings

[17] https://twiki.cern.ch/twiki/bin/view/LHCPhysics/CERNYellowReportPageAt13TeV

[18] https://twiki.cern.ch/twiki/bin/view/LHCPhysics/CERNYellowReportPageBSMAt13TeV

[19] https://twiki.cern.ch/twiki/bin/view/CMSPublic/LumiPublicResults#2015_Proton_Proton_13_TeV_Collis  Lumonisity at CMS