A Passion for Physics

A Passion for PhysicsThe Physics Freshman Seminar

Fall Semester 2008

See information on course web page

Course Organizer/Instructor:Jim Napolitano

Physics Research at RPI

• Physics Department• Astronomy and Astrophysics• Computational Biophysics• Condensed Matter Physics and Optics• Elementary Particle Physics

• Other Departments• Electrical Engineering• Geology (Earth and Environmental Science)• Cognitive Science• And various other things!

No talk summary needs to be handed in for today...... BUT you might want to practice taking notes!!!

Peter Persans (Sep 4)Old Optical Tricks to Study New Materials

Tuning colors with “quantum dots” made from CdSe

Paul Stoler (Sep 11)The Standard Model of Particle Physics

Joel Giedt (Sep 18)Strings and Lattices

Putting “SuperSymmetry” on a lattice.

FIG. 1: The generic mass spectrum of the 5D gravity model showing the heavy first and second

generation scalars and lighter third generation scalars, gluinos, neutralinos and charginos. The LSP

is the gravitino (not shown).

that the Higgs fields obtain the necessary soft masses from this effect, rendering electroweak

symmetry-breaking viable.

Consequently the particle spectrum in our model has very distinctive features and a

generic spectrum is shown in Fig. 1. The first and second generation of scalar partners are

very heavy. These large masses do not destabilize the Higgs mass via radiative corrections

because of (1) small Yukawa couplings and (2) degeneracies at the messenger scale that

prevent large one-loop hypercharge Fayet-Iliopoulos (FI) terms.2 As will be discussed, these

degeneracies are also necessary in order to satisfy flavor changing neutral current (FCNC)

constraints. This spectrum is similar to that considered in Refs. [21, 22] and is also rem-

iniscent of the “more minimal” supersymmetric standard model [6], for which heavy first

two generation scalar fields were considered to ameliorate flavor problems. The LSP is the

gravitino, which means that in our model the lightest neutralino, χ01, is the NLSP. Because

the messenger scale is relatively low, the decay length of χ01 is less than 1 mm. This leads to

a 2γ + !ET (two hard photons and missing transverse energy) signal at the LHC. Although

2 Here it is important that the messenger scale M in our model is O(100) TeV, so that splittings that would

disturb the degeneracies are not introduced under renormalization group evolution to the electroweak

scale.

4

Sang-Kee Eah (Sep 25) Monodispersed Nanoparticles

Gold particles 7nm in size(One gold atom about 0.3nm)

“Bubble Raft”A possible model for self-assembly of nanoparticles

Wayne Roberge (Oct 2)Shock Waves: Things That Go Bump in Space

How do stars(and “solar systems”) form?

Leo Schowalter (Oct 9)Ultraviolet LED’s ... or

My Excellent Adventure Starting a New Business

Providing new technology for an international need ...and being an entrepreneur while doing it.

Mimi Katz (Oct 16)From Phytoplankton to Mammals:

Evolutionary Trajectories linked via the C-Cycle

Ken Connor (Oct 23)Plasma Physics

Using Heavy Ion Beams to Probe Fusion Plasmas.

Selmer Bringsjord (Oct 30)Since Physicalism is False, We Need More than Physics

“SharkTacos.com”

Bringsjord and Zenzen

Heidi Newberg (Nov 6)Your Textbook is Wrong about the Milky Way

If you look carefully, you’re liable to see surprising new things.

Gyorgy Korniss (Nov 13)Dynamics in Networks

We see “Networks” everywhere. What are their common properties?

about 5.25 (Fig. 3A) while !u(")#NPEis re-

duced by only about 1.6% (Fig. 3B). One canalso observe the clear self-averaging propertyfor both global observables, the width and theutilization (Fig. 3). From a broader statisticalphysics viewpoint, one can ask whether asmall, nonzero fraction q of the PEs withrandom links (checked, e.g., at every step) issufficient to control the width. Although thischoice clearly weakens load balancing andthe utilization (Fig. 3B), our results in Fig.3A and recent work on the closely relatedXY-model on a small-world network (25)suggest that a finite width is achieved. Thereis growing evidence that systems without in-herent frustration exhibit mean-field charac-teristics when the original short-range inter-action topology is modified to a small-worldnetwork (23, 25–27).

The generalization when random links areadded to a higher-dimensional underlying reg-ular lattice is clear: Because the one-dimensional case with random links is gov-erned by the mean-field equation, in higherdimensions it will be even more so [i.e., thecritical dimension (18) of the model with ran-dom links is less than 1]. The generalization forthe many-sites-per-PE case also follows fromuniversality arguments: Without the randomconnections, there is an additional fast-rough-ening phase for early times when the evolutionof the time horizon corresponds to random dep-osition (18). Subsequently, it will cross over tothe KPZ growth regime and finally saturate.With many sites per PE, the typically desiredand efficient way of implementing PDES (7,15), the saturation value of the width can be-come extremely large. This underscores the

importance of implementing the additional syn-chronizations through random links to suppressthe roughness of the time horizon.

References and Notes1. R. Fujimoto, Commun. ACM 33, 30 (1990).2. D. M. Nicol, R. M. Fujimoto, Ann. Oper. Res. 53, 249(1994).

3. B. D. Lubachevsky, Bell Labs Tech. J. 5 (April–June),134 (2000).

4. A. G. Greenberg et al., in Proceedings of the 8thWorkshop on Parallel and Distributed Simulation(PADS’94), Edinburgh, UK, 1994 (Society for Com-puter Simulation, San Diego, CA, 1994), pp. 187–194.

5. E. Deelman, B. K. Szymanski, T. Caraco, in Proceedingsof the 28th Winter Simulation Conference (Associa-

tion for Computing Machinery, New York, 1996), pp.1191–1198.

6. D. M. Nicol, in Proceedings of the SCS Multiconfer-ence on Distributed Simulation (Society for ComputerSimulation, San Diego, CA, 1988), vol. 19, pp. 141–146.

7. G. Korniss, M. A. Novotny, P. A. Rikvold, J. Comput.Phys. 153, 488 (1999).

8. G. Korniss, C. J. White, P. A. Rikvold, M. A. Novotny,Phys. Rev. E 63, 016120 (2001).

9. P. M. A. Sloot, B. J. Overeinder, A. Schoneveld, Com-put. Phys. Commun. 142, 76 (2001).

10. D. R. Jefferson, ACM Trans. Prog. Lang. Syst. 7, 404(1985).

11. P. Bak, C. Tang, K. Wiesenfeld, Phys. Rev. Lett. 59, 381(1987).

12. A. G. Greenberg, S. Shenker, A. L. Stolyar, Perfor-mance Eval. Rev. 24, 91 (1996).

13. K. M. Chandy, J. Misra, Commun. ACM 24, 198(1981).

14. B. D. Lubachevsky, Complex Syst. 1, 1099 (1987).15. !!!! , J. Comput. Phys. 75, 103 (1988).16. G. Korniss, Z. Toroczkai, M. A. Novotny, P. A. Rikvold,Phys. Rev. Lett. 84, 1351 (2000).

17. G. Korniss, M. A. Novotny, Z. Toroczkai, P. A. Rikvold,in Computer Simulation Studies in Condensed MatterPhysics XIII, D. P. Landau, S. P. Lewis, H.-B. Schuttler,Eds., vol. 86 of Springer Proceedings in Physics(Springer-Verlag, Berlin, 2001), pp. 183–188.

18. A.-L. Barabasi, H. E. Stanley, Fractal Concepts in Sur-face Growth (Cambridge Univ. Press, Cambridge,1995).

19. T. Halpin-Healy, Y.-C. Zhang, Phys. Rep. 254, 215 (1995).20. M. Kardar, G. Parisi, Y.-C. Zhang, Phys. Rev. Lett. 56,889 (1986).

21. Z. Toroczkai, G. Korniss, S. Das Sarma, R. K. P. Zia,Phys. Rev. E 62, 276 (2000).

22. D. J. Watts, S. H. Strogatz, Nature 393, 440 (1998).23. M. E. J. Newman, J. Stat. Phys. 101, 819 (2000).24. Here we define the Fourier transform of the surfacefluctuations as

$k % ! j%1

NPE

[exp(&ikj)]($j ! $)

where k % 2'n/NPE and n % 0, 1, 2, . . ., NPE – 1.25. B. J. Kim et al., Phys. Rev. E 64, 056135 (2001).26. A. Barrat, M. Weigt, Eur. Phys. J. B 13, 547 (2000).27. M. Gitterman, J. Phys. A 33, 8373 (2000).28. We thank G. Istrate and Z. Racz for discussions. Sup-ported by NSF grants DMR-0113049 andDMR-9981815, Research Corporation grant RI0761, andU.S. Department of Energy grant W-7405-ENG-36( Z.T.).

15 October 2002; accepted 12 December 2002

A Gate-Controlled BidirectionalSpin Filter Using Quantum

CoherenceJ. A. Folk,1,2* R. M. Potok,1 C. M. Marcus,1* V. Umansky3

We demonstrate a quantum coherent electron spin filter by directly measuringthe spin polarization of emitted current. The spin filter consists of an openquantum dot in an in-plane magnetic field; the in-plane field gives the two spindirections different Fermi wavelengths resulting in spin-dependent quantuminterference of transport through the device. The gate voltage is used to selectthe preferentially transmitted spin, thus setting the polarity of the filter. Thisprovides a fully electrical method for the creation and detection of spin-polarized currents. Polarizations of emitted current as high as 70% for both spindirections (either aligned or anti-aligned with the external field) are observed.

The ability to controllably generate and detect electron spin in mesoscopic systems isimportant for the field of spintronics and is

also a powerful tool for investigating basicproperties of spin in coherent electronic sys-tems. The development of tunable spin filters

Fig. 3. (A) Average steady-state width of the virtual time horizon for various values of p for theone-site-per-PE PDES scheme. In addition to ensemble averages over 10 realizations of the randomlinks (solid symbols), a single realization is also shown (the same open symbols). The solid straightline represents the asymptotic one-dimensional KPZ power-law divergence with roughness expo-nent ( % 1⁄2 for the p % 0 case. Note the log-log scales. The q % 0.10 data set corresponds to thecase when only 10% of the PEs have random links and those are checked at every step. (B) Thesteady-state utilization (fraction of nonidling PEs) for the same cases as in (A).

R E P O R T S

www.sciencemag.org SCIENCE VOL 299 31 JANUARY 2003 679

Shengbai Zhang (Nov 20)Physics and Energy Research

“Solving urgent national needs in sustainable energy, environment, technology, and security, through quantum mechanical computational discoveries.”

Angel Garcia (Dec 4)Calculating What Makes Things Alive

Computational biology from a physics perspective.

The EndDon’t forget: No report needed for this talk, but email the grader to make sure you’re able to get through!

See you next Thursday, if not before.