betaFeedback

WorksPhysics and Astronomy

BACKGAMMON: A Scheme for Producing High Intensity Muon Beams for Future Colliders and Other Applications

Abstract

We present a scheme for producing intense μ+μ^+ and μ−μ^- lepton beams that could be utilized in a future muon-muon or muon-ion collider, as well as for other applications. The scheme makes use of BACKscattered GAMMas On Nucleons (BACKGAMMON). Current accelerator infrastructures could be utilized to produce the muon beams, such as the Electron-Ion Collider (EIC) at Brookhaven National Laboratory in the United States. We discuss the implementation of BACKGAMMON at the EIC.

Introduction

There has been growing interest in high intensity muon beams for a variety of applications, such as a muon-muon collider [1], or perhaps a muon-ion collider at Brookhaven National Laboratory in the United States [2, 3]. One approach being studied uses a proton driver, which employs an intense proton beam that impinges upon a target, such as graphite, thereby producing pions, which decay into ∼1014\sim10^{14} muons/s [4]. An alternative approach, called the Gamma Factory, uses high energy photons resulting from resonant backscattering of laser photons off atomic beams of partially stripped ions circulating in CERN’s Large Hadron Collider storage rings [5]. The high energy photons in turn strike a graphite target to produce pions whose decay should also be able to achieve ∼1014\sim10^{14} muons/s. Yet another approach, called BACKscattered GAMMas On Nucleons (BACKGAMMON) [6], and which we present here, uses high energy photons, which are derived from Compton backscattered laser pulses from an intense electron beam, that similarly impinge upon a stationary target to produce pions, which decay into ∼1014\sim10^{14} muons/s. In the latter approach, it may be possible to utilize an existing infrastructure as the electron source, such as the electron storage ring at Brookhaven National Laboratory’s Electron-Ion Collider in the United States. This could be the next phase of research at that facility. As for BACKGAMMON, it was proposed some time ago as a scheme for scattering ultra high energy backscattered photons on targets for a variety of studies, including CP violation in B-meson decay [6], polarized τ\tau decay [7], and various other applications to heavy flavor physics [8].

Backscattered Photon Energy

The analysis of the electron-photon scattering is considered in the laboratory frame, wherein both the electron and low energy laser photon beams are propagating in opposite directions. For an excellent discussion, see Ref. [9]. The general case is depicted in Figure 1, where p1p_1, k1k_1, k2k_2 are the 4-momenta of the incident electron, incident photon, and outgoing scattered photon, respective; and ϕ1\phi_1 is the angle between the incident directions of the electron and photon beams, ϕ2\phi_2 is the angle between the incident electron and scattered photon beams, and θ\theta is the angle between the incident and scattered photon directions.

Schematic of Compton Backscattering

Figure 1. Schematic of Compton Backscattering

For our case, the photon beam is backscattered with θ∼π\theta\sim\pi, and the resulting high energy backscattered photon beam is confined within a cone of opening angle ∼1γ\sim\frac{1}{\gamma} relative to the electron’s incident direction of motion, with γ\gamma being the usual Lorentz contraction factor for the electron beam. We take the incident electron to have energy EeE_e and incident photon to have energy ω1\omega_1, with the electron rest energy being E0E_0. The scattered photon energy ω2\omega_2 is given by

ω2=ω1(1−βcos⁡ϕ1)1−βcos⁡ϕ2+ω1Ee(1−cos⁡θ)(1)\omega_2 = \frac{\omega_1(1-\beta\cos\phi_1)}{1-\beta\cos\phi_2+\frac{\omega_1}{E_e}(1-\cos\theta)} \tag*{(1)}

where as usual, β=vc\beta= \frac{v}{c}; When ω1≪Ee\omega_1 \ll E_e, as in our case, we have the approximation

ω2≈4γ2ω11+4γ2ω1Ee(2)\omega_{2} \approx\frac{4\gamma^{2}\omega_{1}}{1+4\gamma^{2}\frac{\omega_{1}}{E_{e}}} \tag*{(2)}

Unpolarized Total Cross Section

Following Landau, Lifshitz, Berestetskii and Pitaevskii [10], we set

x1=2γω1E0(1−βcos⁡ϕ1).(3)x_{1}=2\gamma\frac{\omega_{1}}{E_{0}}(1-\beta\cos\phi_{1}). \tag*{(3)}

The total unpolarized Compton cross section σCunp\sigma_{C}^{unp} is then

σCunp=2πr021x1[(1−4x1−8x12)ln⁡(1+x1)+12+8x1−12(1+x1)2],(4)\sigma_{C}^{unp}=2\pi r_{0}^{2}\frac{1}{x_{1}}\left[\left(1-\frac{4}{x_{1}}-\frac{8}{x_{1}^{2}}\right)\ln(1+x_{1})+\frac{1}{2}+\frac{8}{x_{1}}-\frac{1}{2(1+x_{1})^{2}}\right], \tag*{(4)}

where r0=2.81794×10−15 mr_{0}=2.81794 \times10^{-15}\ \mathrm{m} is the classical electron radius.

Total Cross Section with Polarization Included

To include polarization effects, one write can write [11]

σC=σCunp+2λePcσ1,(5)\sigma_{C}=\sigma_{C}^{unp}+2\lambda_{e}P_{c}\sigma_{1}, \tag*{(5)}

where λe\lambda_{e} is the mean electron helicity (∣λe∣≤12|\lambda_{e}| \le\frac{1}{2}), PcP_{c} is the Stokes parameter that gives the degree of photon circular polarization given by

Pc=(IL−IR)I0,(6)P_{c}=\frac{(I_{L}-I_{R})}{I_{0}}, \tag*{(6)}

with ILI_{L}, IRI_{R}, and I0I_{0} being the left circularly polarized, right circularly polarized, and total photon intensities. Finally,

σ1=2σ0x1[(1+2x1)log⁡(1+x1)−52+11+x1−12(1+x1)2](7)\sigma_{1}=\frac{2\sigma_{0}}{x_{1}}\left[\left(1+\frac{2}{x_{1}}\right)\log(1+x_{1})-\frac{5}{2}+\frac{1}{1+x_{1}}-\frac{1}{2(1+x_{1})^{2}}\right] \tag*{(7)}

with σ0=π(e2mec2)=2.5×10−29 m2\sigma_0 = \pi\left(\frac{e^2}{m_e c^2}\right) = 2.5 \times10^{-29}\ \mathrm{m}^2.

Luminosity of Electron-Photon Collisions

We use the Suzuki formula [12], where the luminosity

L=NeNphfcos⁡(ϕ2)2π1(σye2+σyph2)(σxe2+σxph2)cos⁡2(ϕ2)+(σze2+σzph2)sin⁡2(ϕ2)(8)L = N_e N_{ph} f \frac{\cos\left(\frac{\phi}{2}\right)}{2\pi} \frac{1}{\sqrt{\left(\sigma_{ye}^2+\sigma_{yph}^2\right)\sqrt{\left(\sigma_{xe}^2+\sigma_{xph}^2\right)\cos^2\left(\frac{\phi}{2}\right)+\left(\sigma_{ze}^2+\sigma_{zph}^2\right)\sin^2\left(\frac{\phi}{2}\right)}}} \tag*{(8)}

where NeN_e is the number of electrons in a bunch, NphN_{ph} is the numbers of photons in a laser pulse, ff is the electron-photon bunch collision frequency, ϕ\phi is the crossing angle of the lines defined by the incident electron and photon momenta, and the σ\sigma’s define the rms sizes of electron and photon beam dimensions at the interaction point, with zz being the longitudinal direction according to Suzuki’s notation. The electron rms beam sizes are given by the Electron-Ion Collider (EIC) Design parameters [13], the laser pulse transverse sizes are taken to correspond to those of the electron bunches, while the laser pulse length is defined by the laser system discussed below.

From the luminosity LL, the rate RR of backscattered photons is given by

R=dNdt=σCL,(9)R = \frac{dN}{dt} = \sigma_C L, \tag*{(9)}

where NN is the number of backscattered photons.

Laser Parameters

The frequency ff at which electron bunches come into the interaction point with the laser pulses is equal to the number of electron bunches in the storage ring (1,160) times the revolution frequency of the electrons around the circumference (3,833.94 meters), or

f=βc11603833.94=90.7 MHz.(10)f = \beta c \frac{1160}{3833.94} = 90.7\ \mathrm{MHz}. \tag*{(10)}

Müller et al. have reported developing a laser system centered at 1046 nm wavelength λ\lambda, 254 fs pulse duration, 80 MHz pulse repetition rate, and 10.4 kW average output power PP [14]. We note that this is not quite at the 90.7 MHz repetition rate of the electron bunches at the EIC; however, they are close enough that we take the repetition rate of both the electron bunches and laser pulses to be 90.7 MHz. For this laser system, the energy per laser pulse is 114.657 μJ\mu\mathrm{J} and the number of incident photons per pulse is 6.03×10146.03 \times10^{14}. The authors report that the average power can be pushed to 100 kW, which would increase the number of photons per pulse by an order of magnitude.

Rate of Backscattered Photons

Table 1 contains updated EIC parameters and the laser system modified to 90.7 MHz pulse rep rate.

NeN_eNphN_{ph}EeE_eω1\omega_1ϕ1\phi_1ϕ2\phi_2θ\thetaϕ\phi
1.72×10111.72 \times10^{11}6.03×10146.03 \times10^{14}4.5 GeV4.5\ \mathrm{GeV}1.19 eV1.19\ \mathrm{eV}π\pi00π\pi00
λe\lambda_ePcP_cϵx\epsilon_xϵy\epsilon_yβx∗\beta_x^*βy∗\beta_y^*ff
12\frac{1}{2}1120 nm20\ \mathrm{nm}1.2 nm1.2\ \mathrm{nm}1 cm1\ \mathrm{cm}1 cm1\ \mathrm{cm}90.7 MHz90.7\ \mathrm{MHz}

Table 1. Input Parameters

The number of photons per laser pulse NphN_{ph} is calculated from the laser system as

Nph=(Pf)(hcλ).(11)N_{ph} = \frac{\left(\frac{P}{f}\right)}{\left(\frac{hc}{\lambda}\right)} . \tag*{(11)}

The laser is focused so that its transverse rms pulse sizes σxph\sigma_{xph} and σyph\sigma_{yph} match those of the electron bunches, which are

σxe=ϵxβx∗(12)\sigma_{xe} = \sqrt{\epsilon_x\beta_x^*} \tag*{(12)}
σye=ϵyβy∗,(13)\sigma_{ye} = \sqrt{\epsilon_y\beta_y^*} , \tag*{(13)}

with ϵx\epsilon_x, ϵy\epsilon_y being the electron beam emittances and βx∗\beta_x^*, βy∗\beta_y^* the electron accelerator lattice functions at the interaction point with the laser beam, giving the values shown in Table 2.

σxe\sigma_{xe}σxph\sigma_{xph}σye\sigma_{ye}σyph\sigma_{yph}σze\sigma_{ze}σzph\sigma_{zph}
14.14 μm14.14\ \mu\mathrm{m}14.14 μm14.14\ \mu\mathrm{m}3.46 μm3.46\ \mu\mathrm{m}3.46 μm3.46\ \mu\mathrm{m}11.0 mm11.0\ \mathrm{mm}76.1 μm76.1\ \mu\mathrm{m}

Table 2. Electron and Laser Beam Sizes

where σze\sigma_{ze} is taken from updated EIC parameters, and the photon pulse length is given by

σzph=c×254 fs.\sigma_{zph} = c \times254\ \mathrm{fs}.

Finally, we obtain Table 3 for the derived backscattered photon energy ω2\omega_2, total Compton cross section σC\sigma_C, luminosity LL, and production rate RR of backscattered photons for Ee=4.5E_e = 4.5 GeV.

ω2\omega_2σC\sigma_CLLRR
340 MeV6.05×10−256.05 \times10^{-25} cm2^21.53×10391.53 \times10^{39}/cm2^2/s0.93×10150.93 \times10^{15}/s

Table 3. Derived Quantities (Ee=4.5E_e = 4.5 GeV)

Muon Production

Tables 1-3 apply essentially to the current state of the electron and laser systems. The rate of muon production from the current systems is a couple of orders of magnitude less than the goal of ∼1014\sim10^{14} muons/s. However, there are several possibilities of enhancing the input parameters. First, [14] states that it is possible to increase the laser average power by an order of magnitude up to 100 kW. Also, further research is needed to achieve another order of magnitude from the number of electrons in a bunch NeN_e, accelerator lattice parameters βx\beta_x, βy\beta_y, and/or electron beam emittances ϵx\epsilon_x, ϵy\epsilon_y. Thus, we can increase the rate RR of backscattered photon production by two orders of magnitude to 101710^{17}/s. In terms of implementing BACKGAMMON at the EIC, there is at least a decade of time to research these issues before it will be possible there, due to the new EIC just starting its research program.

Considering the above comments, to produce the muon beam, 101710^{17}/s backscattered photons of energy 340 MeV impinge upon a stationary target, which we take to be graphite. After varying the target length and radius, we choose the length to be 30 cm and radius 2.5 cm. Both pions and muons are produced, where the pions subsequently decay into muons. The production of muons through the pion channel yields several orders of magnitude more muons than the direct muon channel. A major advantage of the pion channel is that a 340 MeV backscattered photon scattering off a stationary proton (neutron) yields an invariant mass of 1232 MeV (1234 MeV), which sits right at the Δ(1232)\Delta(1232) resonance, thereby boosting the photo-absorption cross section of the process [5]. An added benefit of producing muon beams using the BACKGAMMON scheme is that the numbers of positive and negative muons are comparable. This should be a tremendous advantage for a μ+\mu^{+}-μ−\mu^{-} collider.

Pion Capture

For energies E∼340E \sim340 MeV, the main process leading to γ\gamma beam absorption in matter is the formation of electron-positron pairs. The electromagnetic cascade is formed inside the target, producing secondary photons, electrons, and positrons. Only the small fraction of the photons produce pions. The production and collection of pions and their decay product muons must be optimized in order to produce and collect large numbers of muons.

The method of producing muons used in this paper closely follows the proposed scheme of the Gamma-Factory in Ref. [5]. The Gamma Factory uses a primary photon beam of energy up to 300 MeV, while here we use a photon beam of energy 340 MeV.

Following the approach taken in Ref. [5], we use two methods of selecting pions against the dominant contributions of electrons and positrons. The first method uses the fact that the electrons and positrons have velocities close to the speed of light (βe≈1\beta_e \approx1), while the pions, carrying kinetic energy comparable to their mass, are non-relativistic (βπ≪1\beta_\pi\ll1). The second method exploits the differences in their production mechanisms, leading to their distinct kinematical characteristics. Particularly, the pions are produced with a larger angle with respect to the primary photon beam compared to the electrons and positrons [5]. Thus, the particles having transverse momenta higher than 30 MeV/cc and collected in the angular region θ≥40∘\theta\ge40^\circ are predominantly pions.

Simulations

Simulations have been performed using GEANT4.11.2. Comparisons were made with the earlier version, GEANT4.10.6, that showed good consistency between the two versions. In Figure 2, transverse momentum (2a), total momentum (2b), and kinetic energy (2c) distributions of positive and negative pions 1 mm from the end of the target are shown for the optimal target material, which is graphite [5]. The optimised target length is 30 cm with a radius of 2.5 cm, within the selected pion collection angular region of 40∘40^\circ–120∘120^\circ and transverse momentum greater than 30 MeV/cc. For a photon rate of 101710^{17}/s and beam energy 340 MeV, one can achieve pion fluxes greater than 101310^{13}/s for both positive and negative pions. The pions are produced over a narrow range of momentum and transverse-momentum distributions. The widths of the transverse and total momentum distributions are, respectively, σpT=50\sigma_{pT} = 50 MeV/c and σptot=27\sigma_{p_{\mathrm{tot}}} = 27 MeV/c.

Pion transverse momentum, total momentum, and kinetic energy distributions

Figure 2. a. Pion Transverse Momentum Distribution b. Pion Total Momentum Distribution c. Pion Kinetic Energy Distribution. Pion Kinematic Distributions 1 mm from the end of the graphite target (Photon energy=340 MeV, Photon rate=101710^{17}/s, Target length=30 cm, Target radius=2.5 cm, pion pT≥30p_T \ge30 MeV/c, pion collection angular region is 40∘40^\circ–120∘120^\circ)

Figure 3 shows the transverse momentum, total momentum, and kinetic energy distributions of muons coming from pions decaying before reaching 30 m from the center of the target.

Muon transverse momentum, total momentum, and kinetic energy distributions

Figure 3. a. Muon Transverse Momentum Distribution b. Muon Total Momentum Distribution c. Muon Kinetic Energy Distribution. Muon Kinematic Distributions 30 m from the center of the graphite target (Photon energy = 340 MeV, Photon rate = 101710^{17}/s, Target length = 30cm, Target radius = 2.5 cm, muons resulting from pions decaying before reaching 30 m from center of target, pion pT≥30p_T \ge30 MeV/c, pion collection angular region is 40∘40^\circ–120∘120^\circ)

The plots are made assuming that all the initial pions satisfy the pion selection criteria specified above.

Finally, Figure 4 gives the rate of muon production vs. pion propagation length from the center of the graphite target, yielding ∼1014\sim10^{14} muons/s, which was the goal.

Rate of Muon Production vs. Pion Propagation Length from the center of the graphite target

Figure 4. Rate of Muon Production vs. Pion Propagation Length from the center of the graphite target

The presented studies can be considered as an initial, exploratory step. Conceptual and technical designs of the pion and muon collection scheme needs to be developed. Such studies have been performed for the proton beam-driven muon sources, as outlined in Ref. [15]. The idea can also be implemented for a photon beam-driven muon source.

An ultra high-field solenoid can extend the length of a target upon which a photon beam impinges. This target solenoid collects pions with a large momentum spread and large angles, and guides them downstream into a long solenoid channel, where they decay to muons. Quantitatively, the effect of the solenoidal field profile on the capture efficiency can be evaluated by calculating the muon yield, which is defined as the number of particles that fall within a reference acceptance. A dedicated study of the pion capture system is required.

To conclude, we have presented a scheme called BACKGAMMON, which has the possibility to produce ∼1014\sim10^{14} muons/s for muon-muon and muon-ion colliders, as well as for other applications. One big challenge is the circulating electron beam lifetime, which is on the order of 20 to 200 ms. However, with future research innovations, this problem can be overcome. An example of a revolutionary method of replenishing circulating beams is the Swap-Out mode that was invented at the Advanced Photon Source at Argonne National Laboratory in the United States [16]. It uses fast kicker magnets to extract and dump the stored electron bunches while replacing them with fresh bunches from the injector, all within a few nanoseconds. This process maintains the beam current at a constant value. Indeed, there is considerable research that needs to be pursued to make BACKGAMMON a reality; however, there is a decade or more before Brookhaven National Laboratory can even begin to implement it.

Acknowledgements

The authors would like to thank the following persons for their many insightful comments: J. Scott Berg, Ethan Cline, Abhay Deshpande, Mieczyslaw Witold Krasny, Christoph Montag, and Vadim Sajaev. The study was partly funded by National Science Foundation Grant No. NSF PHY-2310078 at Michigan State University’s Facility for Rare Isotope Beams.

The authors of this paper were ordered alphabetically.

References

  1. [1]C. Accettura et al., Interim report for the International Muon Collider Collaboration (IMCC), https://arxiv.org/abs/2407.12450v1 (2024).
  2. [2]D. Acosta and W. Li, Nucl. Instr. and Meth. in Phys. Res., A1027, 166334 (2022).
  3. [3]E. Cline, J.C. Bernauer, and A. Deshpande, A Future Muon-Ion Collider at Brookhaven National Laboratory, DIS2022: XXIX International Workshop on Deep-Inelastic Scattering and Related Subjects, https://zenodo.org/records/7268146 (2022).
  4. [4]X. Ding, H.G. Kirk, and K.T. McDonald, Particle Production of a Graphite Target System for the Intensity Frontier, 6th International Particle Accelerator Conference, https://accelconf.web.cern.ch/ipac2015/papers/wepje010.pdf (2015).DOI
  5. [5]M. Krasny, The Gamma Factory proposal for CERN, https://arxiv.org/abs/1511.07794 (2015); A. Apyan, M. Krasny, and W. Placzek, Phys. Rev. Accel. Beams 26, 083401 (2023).
  6. [6]S. Mtingwa and M. Strikman, Phys. Rev. Lett. 64, 1522 (1990).
  7. [7]S. Mtingwa and M. Strikman, Nucl. Inst. and Meth. in Phys. Res. A472, 189-194 (2001).
  8. [8]S. Mtingwa and M. Strikman, Nucl. Inst. and Meth. in Phys. Res. A455, 50-53 (2000).
  9. [9]I. Chaikovska, Polarized position sources for the future colliders, Université Paris Sud - Paris XI, https://theses.hal.science/tel-00795937 (2013).DOI
  10. [10]L. Landau, E. Lifshitz, V. Berestetskii, and L. Pitaevskii, Quantum Electrodynamics, Vol. 4, Chap. 10, First published 1971, Latest ed. 2020.
  11. [11]For an excellent discussion, see I.F. Ginzburg, G.L. Kotkin, S.L. Panfil, V.G. Serbo, and V.I. Telnov, Nucl. Instr. and Meth. in Phys. Res. 219, 5-24 (1984).
  12. [12]T. Suzuki, General Formulae of Luminosity for Various Types of Colliding Beam Machines, KEK Pub. 76-3 (1976).
  13. [13]https://www.bnl.gov/eic/.
  14. [14]M. Müller, C. Aleshire, A. Klenke, E. Haddad, F. Légaré, A. Tünnermann, and J. Limpert, Opt. Lett. 45, no. 11, 3083 (2020).
  15. [15]R.B. Palmer et al., “Monte Carlo simulations of muon production,” AIP Conf. Proc. 352, 108–124 (1996).
  16. [16]L. Emery and M. Borland, Possible Long-term Improvements to the APS, in Proc. of PAC’03, Portland, OR, USA, 256 (2003).DOI

Paper details

vlnc:W_rgnevtpb
Version
v1 · latest
Published
Fresh — added to the Hub within the last 30 days
Added
License
CC BY 4.0
Usage

views · 0

downloads · 0

A view counts once per browser per day, a download each time the PDF is requested.

  • Physics and Astronomy

Not recorded for this paper

View statement
Looking for related papers…
Contents