dedup-isc-ft-v107-score
float64
0.32
1
uid
stringlengths
32
32
text
stringlengths
0
32.5k
paper_id
stringlengths
1
14
original_image_filename
stringlengths
5
222
0.430152
f413a7a79ee84f059de6045dd73ad2e0
a) Price P(t) of (<ref>) as a function of time t for α = 0.2. b) Cash C(t) solution of (<ref>) for the case α = 0.2, r=0.1, C_0=10 and d=0.02. c) Similar to curve b) but with d=0.08. d) similar to c) but with C_0=5.
cond
mat0308548-figure2.jpg
0.451136
3a720bb5b8094eb585727f2c3445e80b
a) Price P(t) of (<ref>) as a function of time t for α = -0.2. b) |n|P(t) for the case α = -0.2, r=0.1, C_0=10 and d=-0.08. c) Wealth, W(t) as a solution of (<ref>) and (<ref>) for the parameters of curve b). Same as curve b) except for signs of α, d, α = 0.2 and d=0.08.
cond
mat0308548-figure3.jpg
0.457517
52eccc7311f549be839cb40aff19c41b
Fat solid line: price P(t) from one configuration of the $-game with N=20 agents, r=10%, d_0=8% (assuming dividend d(t)=d_0 P(t)). The parameter values used were s=4, m=8, C_0=50. Stochasticity was introduced via N additional “noise” agents. The fraction of agents allowed to take short positions, ρ =0. Liquidity parameter λ=0.0025. Thin dotted lines represent the 5%, 50% and 95% quantiles (from bottom to top) respectively, i.e. at every time t out of the 1000 different initial configurations only 50 got below the 5% quantile line and similarly for the other quantile lines.
cond
mat0308548-figure4.jpg
0.51237
d11ae6fb79514ead981571e004038fb7
Same as figur 4 but with ρ=0.2
cond
mat0308548-figure5.jpg
0.408391
6754dc39146a4a279bd2cef403c0594e
Same as figur 4 but with ρ=0.4
cond
mat0308548-figure6.jpg
0.496313
fb7ac2dde4004ca4967eb4d92fdd18f2
Fat solid line represents the Nasdaq Composite index. Dotted line represents the fraction of change in shares shorted, n_s(t), defined as the monthly total number of change in shares shorted divided by the total monthly share volume. The dashed line represents the cumulative fraction of change in shares shorted, Q_s(t) = ∑_0^t n_s(t). Both n_s(t) and Q_s(t) have been multiplied with a factor 10^5.
cond
mat0308548-figure7.jpg
0.527081
ee3c6b4a77ec40bdb5747f758bed4264
Linear conductance of an N-AB-S junction (see Fig. <ref>) at T=0 and zero magnetic field as function of level position ϵ at γ=0.1 (solid line) and γ=0(dash line)
cond
mat0308551-fig2.jpg
0.477285
16288e8691c348a19aee51dd76c652d7
Linear conductance of an N-AB-S junction (see Fig. <ref>) at T=0 and zero magnetic field as function of level position ϵ at γ=0; 0.1 and γ=0.2.
cond
mat0308551-fig3.jpg
0.591307
478158cc24444cf1844a8d3208d35db7
Linear conductance of an N-AB-S junction (see Fig. <ref>) at T=0 versus AB phase for different level positions (ϵ=-0.4 top line, -0.34 middle line, -0.24 bottom (red) line. All curves correspond to γ=0.1. The parameters are tuned to cover the Kondo and crossover to Kondo regimes.
cond
mat0308551-fig4.jpg
0.502498
eb6c8434aec14e5992ccc29d608cc3cf
The stroboscopic map corresponding to a clean trajectory in the Chua circuit.
cond
mat0308554-chua_10.jpg
0.458907
132a45db2b47462d97e913908a4f94cc
The stroboscopic map received after the noise reduction by the LPNC method applied to data presented at Fig. <ref>.
cond
mat0308554-chuacl_12.jpg
0.482956
0137b95a61dd458cbc0b595b2fb2b278
Stroboscopic map received after noise reduction by the LPNC method applied to data presented at Fig. <ref>
cond
mat0308554-chuadncl_14.jpg
0.468604
9c4934b113794f4ebeb9dea0c922c556
Stroboscopic map received from the Chua circuit in the presence of a mixture of a measurement and dynamical noise 𝒩≈ 22%.
cond
mat0308554-chuadnno_13.jpg
0.362605
a42ca8227d964ab2bb929981f11a6c07
The stroboscopic map received from the Chua circuit in the presence of a measurement noise 𝒩=96.5%. Note the difference in scale.
cond
mat0308554-chuano_11.jpg
0.464244
41e9f88ec7954b9dbf6b6b7d40cbba92
The plot of the gain parameter 𝒢 versus number of neighbors of the GHKSS method (squares) and LPNC method (triangles). Lorenz system 𝒩=78%, N=1000.
cond
mat0308554-iterGHKSSLPNC_2.jpg
0.462442
d8e80192d46f435b93a6ae1cf636f1c8
The plot of the gain parameter 𝒢 versus number of iterations of the GHKSS method (squares) and LPNC method (triangles). Lorenz system 𝒩=78%, N=1000.
cond
mat0308554-nnGHKSSLPNC_1.jpg
0.999953
17680b39d6c1432d8031ffabd98fed44
Delaunay triangulation for a set of nine points in a two-dimensional space. The near neighbors are connected by bold lines. Sets T𝐱_i, i=1,2,…,9 are limited by thin lines.
cond
mat0308554-redfig_TD_5.jpg
0.999953
41cea55535bd481a80b69298ce6154bc
Illustration of nearest neighborhood search by DT approach a) 𝐱_j and 𝐱_n are not near neighbors. b) 𝐱_j and 𝐱_n are near neighbors
cond
mat0308554-redfig_TDprz_6.jpg
0.435544
5bc747829b164f3c95be4990db9308e1
The random data from uniform distribution
cond
mat0308554-redfig_szuma_3.jpg
0.464156
7cc0327232e44e8fb3a71324e678d9ad
The random data shown in the Fig. <ref> after noise reduction with the LPNC method
cond
mat0308554-redfig_szumc_4.jpg
0.566124
a17f8448cede43e28762f9d1f752514a
Chaotic Henon map without noise
cond
mat0308554-redfighenon_a_7.jpg
0.458228
e876ee2a14af49c697625029c6b63e04
Chaotic Henon map with a measurement noise 𝒩=69%. Note the difference in scale.
cond
mat0308554-redfighenon_b_8.jpg
0.500023
230ff3ac1f0c4767820b73c28fa3d510
Chaotic Henon map with a measurement noise after noise reduction with the LPNC method, 𝒢=4.3
cond
mat0308554-redfighenon_c_9.jpg
0.507107
75bac38617f848f9a4318145900f3aa8
Free energy vs. temperature for a semiflexible polymer: (a) Flory's calculation; (b) Gujrati-Goldstein upper bound.
cond
mat0308555-Figure1.jpg
0.420926
b1766f47d7ff4510a80d15c774391139
a=0.5,b=0. (a) Entropy in the 2-cycle FP scheme for the ML (continuous line), EL/SCL (dashed line) and CR (dash-dot line). We also show T_CRE(⧫), T_M(∘) and T_MC (∙); (b) magnification of the area contained in the box in (a).
cond
mat0308555-Figure10.jpg
0.499712
4aeee2a215ac41d192333e2d8f03bddb
a=0.5,b=0. Specific heat in the 2-cycle FP scheme for the ML (continuous line), EL/SCL (dashed line) and CR (dash-dot line). We also show T_CRE(⧫), T_M(∘) and T_MC (∙).
cond
mat0308555-Figure11.jpg
0.512085
3f25ad92091f41f4ab214e9df322b483
(a) Gauche bond and (b) parallel bond density in the 2-cycle FP scheme for the ML (continuous line), EL/SCL (dashed line) and CR (dash-dot line). We also show T_CRE(⧫), T_M(∘) and T_MC(∙).
cond
mat0308555-Figure12.jpg
0.485695
f890e258b94647d9a0fc913d53bbbe36
Effect of b on the phase diagram of the system (a=0.5). The free energy of ML and CR does not depend on b. Three EL/SCL curves are shown corresponding to b=-0.4 (long dash), b=0 (medium dash) and b=0.4 (short dash).
cond
mat0308555-Figure13.jpg
0.497998
368d8b31ed6949a7bf8f00a4774334ee
Dependence of the melting temperature on a. The first-order transition line (continuous line ) and the second-order transition line (dashed line) are shown.
cond
mat0308555-Figure14.jpg
0.435505
ace3bdf149814925a040a2d94e1e7243
Upper half of a Husimi cactus of generation m=3. The dangling bonds outside it show its connection through surface sites (not shown in the figure) with the larger infinite cactus, as explained in the text.
cond
mat0308555-Figure2.jpg
0.534313
a09a1247b87b497ea8da3d230362b74b
The possible states of a square in the lattice: (a) no bonds, (b) one bond, (c) two bonds and (d) three bonds.
cond
mat0308555-Figure3.jpg
0.446233
3f9b934224474ce1b7884dbcd6a389dd
Possible configurations of the polymer chain at T=0: (a) crystalline phase with N_g=0 and N_p=N; (b) step-like configuration with N_p=0 and N_g=N. See text for the explanation of the symbols.
cond
mat0308555-Figure4.jpg
0.494253
61d5728905f143428db22a511f20ad25
The four possible states of the polymer chain at any site at the mth level of the lattice.
cond
mat0308555-Figure5.jpg
0.530078
d39421b452834e14966cd6df8f1ac503
Possible configurations of the polymer chain when the mth level site is in the I state.
cond
mat0308555-Figure6.jpg
0.591808
c435d15659264bf78babd43d07fdde88
Possible configurations of the polymer chain at the origin of the tree.
cond
mat0308555-Figure7.jpg
0.446705
9b9d40513d4843169206c304ac4eb57e
a=0.5,b=0. Free energy in the 2-cycle FP scheme for the ML (continuous line), EL/SCL (dashed line) and CR (dash-dot line). We also show T_CRE(⧫), T_M(∘) and T_MC (∙). Here, as well as in Figures 10 to 12, the stable phases are represented by thick lines while the metastable phases are represented by thin lines.
cond
mat0308555-Figure8.jpg
0.417335
336b1ff3b3704456b5893c2e7a63208e
Dependence of l_a and l_b on the temperature for the two phases obtained at low temperature in the 2-cycle fixed point scheme.
cond
mat0308555-Figure9.jpg
0.430125
15ad14e8dd9e40d08093c90ff457a5c8
Schematic of the overgrown corner showing the substrate and precleave 2DES's (left) as well as a top perspective of the contact layout (right).
cond
mat0308557-EP2DS-15-Fig1.jpg
0.457226
497f3f2e77b543ee8a98223d9fd57437
Plot of R_xx^s = R_5-8,6-7 (top), R_xx^p = R_1-4,2-3 (middle), R_xy^s = R_2-6,7-5, R_c = R_2-6,1-5, and R_xy^p = R_2-6,3-1 (bottom) measurements of the corner well at tilted B fields such that ν_s/ν_p = +1/3. n_s = 1.07 × 10^11 cm^-2 and n_p = 1.30 × 10^11 cm^-2. Plateaus in R_xy and R_c are indexed with quantum number, n according to R = h/n e^2. The bottom figure demonstrates the Landauer-Büttiker prediction R_c = |R_xy^s| - |R_xy^p|.
cond
mat0308557-EP2DS-15-Fig2.jpg
0.474691
3aeb23fd3d584a0285c72459981e7a2f
Plot of R_xx^s = R_5-8,6-7 (top), R_xx^p = R_1-4,2-3 (middle), R_c = R_2-6,1-5, R_xy^s = R_2-6,7-5 and -R_xy^p = R_2-6,1-3 (bottom) measurements of the corner well at tilted B fields such that ν_s/ν_p = -1/2. n_s = 1.15 × 10^11 cm^-2 and n_p = 1.25 × 10^11 cm^-2. Plateaus in R_xy and R_c are indexed with quantum number, n according to R = h/n e^2. The bottom figure demonstrates the Landauer-Büttiker prediction R_c = |R_xy^s| + |R_xy^p|.
cond
mat0308557-EP2DS-15-Fig3.jpg
0.422779
8734f5f763234bf0b7614ba0caafae2c
Energy dependence of the magnetic scattering χ"_c(Q,E) at different wave vectors Q measured with polarized neutrons.
cond
mat0308558-Edep.jpg
0.447483
53ae58f941f441268ff3c8c67e26751a
Wave-vector dependence of the incommensurate spin fluctuations at E=6 meV. Open circles are measured near Q=(0.7,0.7,0) <cit.> using unpolarized neutrons, which probes χ_ab+χ_c+non-magnetic scattering. Closed symbols are measured near Q=(0.3,0.3,0) using polarized neutrons, where χ_ab and χ_c are probed separately. In order to obtain the same intensity scale, the polarized data are rescaled with the magnetic form factor and a constant of 118, corresponding to the non-magnetic scattering, is added.
cond
mat0308558-Qdep.jpg
0.437021
e38706bb233e4228ac8cf9d60ee266da
|K_T_1| (=|K^qp|√( K(α)) ) plotted against |K_γ| (=|K^qp|/R ) for various HFS's. |K_T_1| and |K_γ| are estimated from the quasi-particle Korringa relation (T_1T(K^qp)^2 K(α)=constant) and the electronic specific heat (K^qp∝γ_elR) just above T_c, respectively. Here we assume A_hf(q)=A_hf for calculations of K_T_1, and μ_eff≈1.73μ_B for U-based HFS's (for details, see 3). The solid line shows R√( K(α))=1.
cond
mat0308562-fig1vfeps.jpg
0.448642
52f898a1e19e4945b2d8c0b2ad1c8921
Current-voltage characteristic of a quantum dot connected to three ferromagnetic leads i∈{1,2,3}, with respective polarizations P_i, through tunnel junctions with capacitances C_i and net tunneling rates γ_i (circuit shown in the inset). A voltage bias V is applied to leads 1 and 3; lead 2 is connected to ground. The average current I_2 through lead 2 is shown as a function of voltage, for C_1=C_2=C_3, γ _1=γ_2/50=γ_3/10, k_BT/E_0=0.1, and different values of lead polarizations. The current is plotted in units of eγ_tot =eγ_2(γ_1+γ_3)/(γ_1+γ_2+γ_3); the voltage in units of V_0=E_0C/(C_1+C_3)e; E_0 is the position of the dot level. For P_1=P_2=P_3, I_2 coincides with the paramagnetic case (diamonds). In the other cases, the high-voltage limit of I_2 can be larger or smaller than the paramagnetic value, depending on the lead polarizations. For P_1=-P_2=P_3=0.6 (circles), the effect of spin-flip scattering is shown. Spin-flip scattering makes the I_2-V curve tend to the paramagnetic one.
cond
mat0308564-Figure1.jpg
0.410584
d5b6e3b48f914653944dd4d1c598bb73
Fano factor F=S_22/2eI_2 of lead 2 as a function of voltage, for the same circuit parameters as in Fig. 1. In all curves γ_sf=0. For P_1=P_2 =P_3, the Fano factor is different from that of the paramagnetic case (diamonds) in contrast to what happens for the average currents. The inset shows the typical time dependence of the spin on the dot, in the high-voltage limit V≫ V_0 for the case P_1=P_2=P_3=0.6.
cond
mat0308564-Figure2.jpg
0.502678
b3b4f16854e04d7792341a789eb9af68
Current cross-correlations between leads 1 and 3 as a function of voltage. The curves are shown for the same circuit parameters as in Fig. <ref>. The cross-correlations can be positive in the cases P_1=-P_2=P_3=0.6 (circles) and P_1=P_2=P_3=0.6 (squares). Note that the sign of cross-correlations can be reversed just changing the sign of P_1. In all curves γ_sf=0. The inset shows the influence of spin-flip scattering on the cross-correlations in the high-voltage limit V≫ V_0. In the paramagnetic case (diamonds), spin-flip scattering has no effect. In the limit γ_sf≫γ_tot, the cross-correlations tend to the paramagnetic value.
cond
mat0308564-Figure3.jpg
0.532978
97c80a396c6c4387b708e982024b1c80
Influence of asymmetry between γ_2 and γ_1+γ_3 on the high-voltage limit of the cross-correlations, for P_1=P_3=P_3=0.6 (squares) and -P_1=P_3=0.9, P_2=0 (hexagons), for γ_3/γ_1=10 (full symbols) and γ_3/γ_1=1 (empty symbols). Large values of γ_2/(γ_1+γ_3) favor positive cross-correlations. For -P_1=P_3=0.9, P_2=0, an asymmetry between γ_1 and γ_3 is also necessary. The vertical dashed line indicates the ratio γ_2/(γ_1+γ_3) corresponding to Figs. <ref>, <ref>. The two insets show the high-voltage limit of the cross-correlations as a function of P_3, for γ_1=γ _2/50=γ_3/10, P_1=P_3 (left inset) and P_1=-P_3 (right inset) and P_2=0 (dashed lines) or P_2=0.6 (full lines). For all curves γ_sf=0.
cond
mat0308564-Figure4.jpg
0.548649
017f91f2696d476eae2b726ebb90aeb2
Configurations of the Mn_ Ga-Mn_ int pair in the (110) plane of GaAs crystal <cit.>: (a) Mn_ int in T(Ga_3Mn) position is the nearest neighbor of Mn_ Ga; (b) Mn_ int in T(As_4) position representing a partially dissociated pair at a doubled distance; (c) Mn_ int in the T(As_4) position closest to Mn_ Ga.
cond
mat0308568-fig2.jpg
0.397347
223e9fe889954e9590aeef966b6a3816
Spin splitting of the valence band as a function of Mn concentration.
cond
mat0308568-fig3.jpg
0.425899
b8585e228de647479765a3dfb180ac1d
Optimal number of risky assets ϕ in the hedging portfolio, as a function of the level x of the underlying asset for different strategies within the Black-Scholes. As expected the different curves are very similar. The value of the strike is K=110, and the time is half the maturity of the option: k=N/2.
cond
mat0308570-fig.jpg
0.441722
05391b71a99c4547a60204758e144c54
Optimal number of risky assets ϕ in the hedging portfolio, as a function of the level x of the underlying asset for different strategies: Black-Scholes, Leland, Δ_0=-1 and Δ_0=-5, for β=0.05. The value of the strike is K=110.
cond
mat0308570-fitransg.jpg
0.484915
e94ff97670154cf1b8f285fb4c76f6c3
implied volatility smile at t=0. ν=4 and S_0=100
cond
mat0308570-smile.jpg
0.402941
ff6394225c0a4e7084e83223cb4d4b12
Geometry of a single-electron quantum dot on helium surface. A micro-electrode is submerged by the depth h∼ 0.5 μm beneath the helium surface. The electron is driven by a field E_⊥ normal to the surface. This field is a sum of the electrode field ℰ_⊥ and the field of the parallel-plate capacitor (only the lower plate of the capacitor is shown). The confining in-plane potential U__∥( r) [r=(x,y)] is determined by the electrode potential and geometry.
cond
mat0308572-dykman_fig1.jpg
0.5304
bfe120a2afc64dc7b4bc4dd1b5ffa9ef
Imaginary part of the on-site propagator and of the on-site self-energy (inset) in Matsubara frequencies for U=2,2.1,2.2 (green curves with triangles), U = 2.25,2.29,2.31 (blue curves with squares), U= 2.34,2.43,2.5 (red curves with circles) and T/D =1/44.
cond
mat0308577-ImG_Sig.jpg
0.428175
267c82fd5c2749af8bde613443297ca9
The real space cluster and labeling conventions.
cond
mat0308577-cluster.jpg
0.508332
a6127b88fd0545fda6f016832e817f27
Double occupancy as a function of U. The curves correspond (from bottom to top) to T/D =1/20,1/30,1/40,1/44.
cond
mat0308577-dandinset.jpg
0.449861
2ee1c908ac9c46e0822a11c00d99eea3
Zm^*/m as a function of U at β D =44. Top (bottom) curve corresponds to the hot (cold) point. Inset :Σ'_12 vs. ω_n for U=2,2.1,2.2,2.25,2.29,2.31 (from top to bottom) and T/D=1/44.
cond
mat0308577-ratioinset.jpg
0.517975
d181c3bb8f7a4d9984a7867d969c3d60
Switching field versus wire orientation for a 90^∘ bent nanowire. The dots indicate the measured switching fields of the two segments; the open triangles show where only a single transition is observed. The solid lines are the small-nucleation-volume fit from Fig. <ref>, shifted in angle to account for the different orientations of the individual segments.
cond
mat0308579-Silevitch_fig10.jpg
0.452994
24e3757ee4684836b7e91a6d2875f4b5
Resistance vs. wire orientation for a 40 μm nanowire with a 90^∘ bend at fixed external field (a) H=450 Oe. (b) H=1000 Oe. (c) H=3900 Oe. The arrows indicate the direction of rotation. (d) Switching angle vs. applied field. The filled and open circles are switching transitions seen in counter-clockwise (θ increasing) and clockwise (θ decreasing) rotations, respectively. The solid line and circle represent trajectories followed in (H,θ) by Fig. <ref>(b) and in panel (a) of this figure, respectively. The open diamond and square indicate the position of switching events that occur on both trajectories.
cond
mat0308579-Silevitch_fig11.jpg
0.453644
2d8c127213484318b582361a6cc6731f
(a) Atomic force microscopy image of a bent Ni nanowire. (b), (c) Magnetic force microscopy images of the same nanowire. In each panel, the wire was magnetized in the direction indicated by the arrow and subsequently imaged in zero field.
cond
mat0308579-Silevitch_fig12_smaller.jpg
0.448607
97fd9c76f309442c924b494c9e58993a
Schematic of the magnetic configuration of a 90^∘ bent nanowire at two orientations to the field; the angles between the segments and the field are the same in both cases. (a)–(d) Nanowire in “wall” orientation (see text). (e)–(h) Nanowire in “no-wall” orientation. The applied field is given between each row. Note the presence of a domain wall in case c, but not in g.
cond
mat0308579-Silevitch_fig13.jpg
0.442912
cd449a7857774da1a85d1085bde720aa
(a) R(H) of a straight nanowire at 28^∘ (solid) and 62^∘ (dashed) to the field. (b) R(H) of a 90^∘ bent wire, with segments at 28^∘ and 62^∘ to the field.
cond
mat0308579-Silevitch_fig14.jpg
0.482902
3e1f1a9eb9694f5385512b0595c43a6a
Resistance versus field for a 90^∘ wire with segments at 8^∘ and 82^∘ to the field. (a) Solid line is the wire in the “wall” orientation (see text); dashed line is wire in the “no-wall” orientation. (b) Difference in resistance between the two wire orientations.
cond
mat0308579-Silevitch_fig15.jpg
0.450275
0813598390c941ea9d12252fb870ee6f
SEM micrographs of PtNiPt nanowires. (a) Energy-resolved image of 3-segment nanowire showing Pt endcaps and Ni central segment. (b) Nanowire with 90^∘ bend at the center. (c) Detail of bend region for the wire shown in panel (b).
cond
mat0308579-Silevitch_fig1_smaller.jpg
0.47312
b3e54a63bb934283ad2f58f172073cf3
Optical micrographs of PtNiPt nanowires with lithographic electrical contacts. (a) 20 μm long straight nanowire. (b) 35 μm long nanowire with 25^∘ bend.
cond
mat0308579-Silevitch_fig2_smaller.jpg
0.488944
11f36eda6b984a45a2121eae94fc754e
Magnetoresistance of a 20 μm long straight nanowire at different angles θ to the applied field. (a)–(d): R(H) at θ=0^∘, 70^∘, 82^∘, and 88^∘, respectively. The open square and diamond mark switching events at the same (H,θ) locations marked by the corresponding symbols on Fig. <ref>(b). The arrows indicate direction of field sweep.
cond
mat0308579-Silevitch_fig3.jpg
0.494081
5038862675e8433cad4dc86eebf28d89
(a) Polar plot of switching fields for a 20 μm straight wire. The external field is applied in the θ=0 direction; the nanowire points radially outward. Circles: Measured H_sw(θ). Solid line: Curling small-nucleation-volume fit (see text). (b) Data and model from (a), replotted on linear scale, showing |H_sw|. Switching fields from four nanowires with lengths ranging from 12 to 20 μm are plotted. Dashed line: Predicted H_sw for simultaneous reversal of the entire wire. The line drawings at the top of (b) show the direction of the wire orientation relative to the horizontal field.
cond
mat0308579-Silevitch_fig4.jpg
0.492319
d1cbb7cdd1904e2996e9af045d0123ba
Resistance vs. wire orientation at fixed external field (a) H=300 Oe, (b) H=500 Oe, (c) H=1000 Oe for a 20 μm straight nanowire. The arrows show the direction of rotation. The line drawings above plot indicate wire orientation. The open diamond and square in panel (b) indicate switching events at the same (H,θ) locations marked by the corresponding symbols on Fig <ref>(b). (d) Switching angle vs. applied field. The field points in the θ=0 direction and the nanowire points radially outward. The solid circles are from measurements with θ increasing, and open circles with θ decreasing. The solid curve is the curling-mode fit to the switching field data shown in Fig. <ref>. The straight line and circle show the trajectories in (H,θ) for the data in Fig. <ref>(b) and panel (b) of this figure, respectively.
cond
mat0308579-Silevitch_fig5.jpg
0.506562
bd7eef4c5c19460fb977000b2af654d5
Calculation of magnetoresistance and magnetostatic energy in straight nanowire using modified Stoner-Wohlfarth magnetization model. (a)–(d) Straight nanowire at 46^∘ to the field. (a) Measured and model magnetoresistances. Model is a best-fit curve to equations (<ref>) and (<ref>), with H_0=1036 Oe, Δ H=767 Oe, H_sw=260 Oe, R_0=24.67 Ω, and Δ R=0.306 Ω. The model curve has been shifted downwards for clarity. Arrows show the direction of field sweep. (b) ω (H), the angle between the magnetization vector and the wire axis. Dashed lines show magnetization reversal at the curling mode switching field. (c) M(H) for the model curve shown in (a). (d) Magnetostatic energy calculated from (b) and (c). (e)–(h) Straight nanowire at 87^∘ to the field. H_0=5067 Oe, Δ H=2373 Oe, H_sw=1650 Oe.
cond
mat0308579-Silevitch_fig6.jpg
0.510526
2a9fcaba16d543efbfbe941d6746a919
Magnetoresistance of a 35 μm long 25^∘ bent nanowire at different angles to the applied field. (a) Both segments at 12^∘ to H (θ=180^∘ on Fig <ref>). (b) Segments at 41^∘ and 66^∘ to H (θ=126^∘). (c) Segments at 57^∘ and 82^∘ to H (θ=111^∘). (d) Segments at 77^∘ and 78^∘ to H (θ = 90^∘).
cond
mat0308579-Silevitch_fig7.jpg
0.562181
199609cf71fc4a44b0b88bdbf4a0a2ff
Magnetoresistance R(H) of a 35 μm long 90^∘ bent nanowire at different angles to the applied field. (a) Both segments at 45^∘ to H (θ=180^∘ on Fig <ref>). (b) Segments at 20^∘ and 70^∘ to H (θ=155^∘). Open diamond and square mark switching events that occur at the same location as corresponding points marked on Fig. <ref>(b). (c) Segments at 7^∘ and 83^∘ to H (θ=132^∘). (d) Segments at 2^∘ and 88^∘ to H (θ=137^∘).
cond
mat0308579-Silevitch_fig8.jpg
0.475161
d52f7f6fbd1344c28135ffbe5e1d38bc
Switching field versus wire orientation for a 25^∘ bent nanowire. The dots indicate the measured switching fields of the two segments; the open triangles show where only a single transition is observed. The solid lines are the small-nucleation-volume fit from Fig. <ref>, shifted in angle to account for the different orientations of the individual segments.
cond
mat0308579-Silevitch_fig9.jpg
0.608619
33b4a1f1ae1b4f6db191966c5521d949
BN flat sheets under uniaxial strain η_xx (a) and shear strain η_xy (b). In both cases, threefold symmetry is broken and charge redistribution gives rise to a net dipole moment. The corresponding polarization directions (P) are marked by arrows.
cond
mat0308583-fig1.jpg
0.486085
236d6d61239d4b159fcf02d3bec08df4
Schematic structures of (a) stretched (n,0) nanotubes and (b) twisted (n,n) nanotubes where arrows display the strain deformation direction in the tangential plane. Panels (c) and (d) show the calculated tube piezoelectric constant e_11/C and e_14/C as functions of 1/n^2 where C is the tube circumference. Contributions from σ and π electrons to the total piezoelectric response are separated.
cond
mat0308583-fig2.jpg
0.429911
dacd78b57ea24c4fb5309f53653e5237
Piezoelectric response as function of the chiral angle in a sample of chiral nanotubes experiencing the uniaxial strain (a) and the shear strain (b). Solid black curves are the analytical result, Equation (<ref>). Panel c shows ratio of the piezoelectric constants of chiral nanotubes to their flat sheet values plotted as a function of the inverse square radius. The two branches are for the uniaxial e_ 11 and torsional e_ 14 response.
cond
mat0308583-fig3.jpg
0.419257
3f633bd59c22464c9fd99e3657be31fd
Solid line: function Ñ_b^∞(ω) for the (110) direction, calculated using complete phonon dispersions. Dotted line: parabolic approximation to Ñ_b^∞(ω), for a non dispersive medium. Inset: dispersion relations for a 2.2 nm wide wire with frozen boundary.
cond
mat0308587-fig1mingo_rev.jpg
0.483595
60379d2aabab4fa885b1cc1b04006b74
Thermal conductivities versus temperature calculated using the complete dispersions transmission function, for F l=1.05× 37 nm (solid), 1.3× 56 nm (dotted) and 1.15× 115 nm (dashed). Dots: experimental results from ref. Li. Inset: bulk.
cond
mat0308587-fig2mingo_revNEW.jpg
0.413173
0267ea83eb3a46f0b614683a6e14acf0
Thermal conductivity normalized by its value at 320^oK, for nanowires of different widths, calculated by the Callaway formula with cutoff at ω_D (86 THz) (a), and at ω_C (42 THz) (b). (Experimental results from ref. Li.) Inset: bulk thermal conductivity for the two cases (overlapping almost completely) and experimental curve from ref. Holland (dots).
cond
mat0308587-fig3mingo_revNEW.jpg
0.413073
d5d38c95883c4de1a16d3cbad51822f2
Flow of the renormalized minimum of the potential κ_a (solid), the quartic bosonic coupling 10^-2λ_a (long dashes) and the wave function renormalization 10^-1Z_a (short dashes). We have chosen U/t=3 and T/t=0.15; μ=0.
cond
mat0308591-ayuk_ssb.jpg
0.413302
5d8d30b289e24a3ba266ad6132975e68
Renormalized expectation value κ_a of the antiferromagnetic condensate in the low temperature phase for μ=0, U/t=3.
cond
mat0308591-kappa_T.jpg
0.486154
2c6e3d0dcc75455098ccd50dc6c4b53f
Right: edge of the sheet seen from above (polypropylene 27 μm thick) for two diameters of the cylindrical cutting tip: w=31mm and 5mm. (a) Raw experimental snapshots of the cracked film edge and (b) after rescaling according to indenter radius; (c) numerical simulation.
cond
mat0308594-typiquescales.jpg
0.461074
b5ab9888b6a248d08f9a3e0f744611ce
The energy per particle as a function of density for κ=1.
cond
mat0308596-fig1.jpg
0.520046
8fb88e9b7f334ab5b03a99716d6ac3af
Typical dispersion curves for the one (a) and three (b) chain structures for two different magnetic field values.
cond
mat0308596-fig10.jpg
0.471962
358bd7dc4c76438c98ed606a1a2e69ab
The magnetic field dependence of the softening of the phonon mode at the structural transition from one to two-chains.
cond
mat0308596-fig11.jpg
0.516948
7eff3b84ee1544ad8e835d0ed2a9095d
Particle trajectories for 3 × 10^7 MC steps for three different temperatures and three different values of the density at κ = 1.
cond
mat0308596-fig12.jpg
0.477685
acfb3008e713437b82d002013e207137
Particle trajectories for 3 × 10^7 MC steps which qualitatively illustrates the different melting behavior at the boundaries due to the confining potential for κ =1.
cond
mat0308596-fig16.jpg
0.447919
9215665035514bdfacef459c7da4c5c1
The behavior of the Lindemann parameter, for the single-chain regime at four different densities. It shows how the linear regime at higher densities becomes sublinear at lower densities.
cond
mat0308596-fig18.jpg
0.440763
2179c1a79ab4498eb62779956ab7a4e3
The pair correlation function at different temperatures, for three different densities, for the single chain configuration.
cond
mat0308596-fig19.jpg
0.429126
9b017c6549074d659ef2a3a334a91f9b
The zero temperature structural phase diagram.
cond
mat0308596-fig2.jpg
0.514512
61574e238cc04c2f80666bc8a33842aa
The height of the first and second peaks of the pair correlation function for the single chain as a function of temperature for three different densities.The lines are the best fits with the function α (T/T_0)^-β.
cond
mat0308596-fig20.jpg
0.587416
185507d24fdb4849abd125ab3854e8de
The lateral position of the chains in the Wigner crystal state as a function of the linear density for κ=1.
cond
mat0308596-fig3.jpg
0.525381
2b0fa73889ad4d83b157fddb5f896331
The energy per chain at T=0 and (a) κ=0.01, (b) κ=1 and (c)κ=10. The energy is always higher for the external chains but as the Coulomb limit (κ≪ 1) is approached the difference is diminished and the system behaves isotropically.
cond
mat0308596-fig4.jpg
0.457001
0e2ea40048594028a49f12f3b6ee547d
The derivative of the energy with respect to the density for κ =1. Only the transition from one to two-wires is continuous (second order) the rest are first order.
cond
mat0308596-fig5.jpg
0.515498
98200fb9c9b44123912d1ac9f2c6fa33
The mechanism of the structural transitions 1⟶ 2 chains and 2 ⟶ 4 chains.
cond
mat0308596-fig6.jpg
0.430367
bedd13f653da4081b4381d78a8662739
The normal modes of the system in the one (a), two (b) and three (c) chain configuration. The optical and acoustical branches correspond to motion in the confined and unconfined direction respectively. The wavelength is in units of π/a, where a is the length of the unit cell.
cond
mat0308596-fig7.jpg
0.424988
b362f8896fc742009516f7fb59215991
The phonon spectrum at the softening of the optical mode at the structural transition from one to two-chains.
cond
mat0308596-fig8.jpg
0.494553
9ce1572e1cf74fd68640ad506d9854f0
The motion of the particles for the one-, two- and three- chain structure which corresponds to the different eigenfrequencies.
cond
mat0308596-fig9.jpg
0.427498
96411b3732854cdeaa2a78ac668d8cee
The summary for model 2 of the various indicators of critical behavior are shown as a function of the system size. Note that this fit (dashed line) suggests that a finite-temperature phase transition does not survive taking the thermodynamic limit.
cond
mat0308602-Sum.jpg
0.41886
e6141d0c9989410e8370a5354282edd5
The maximum of the susceptibility times the temperature for model 1. The slope of this gives an estimate for the ratio of critical exponents γ/ν.
cond
mat0308602-Wheeler.jpg
0.46994
c1630626bfed42b29502b52c9233d3df
The total number of Ising spins N_ tot as a function of the number of spins N_0 in the one-dimensional chain is shown. The solid line is a fit to the data for N_0≥2^10. Also shown is the standard deviation of N_ tot from the mean (squares), and a fit to those data (dashed line).
cond
mat0308602-aves.jpg