Upload 9401.json with huggingface_hub

9401.json

@@ -0,0 +1,42 @@
+{
+    "9401/astro-ph9401015_arXiv.txt": {
+        "abstract": "The angular power spectrum estimator developed by Peebles (1973) and Hauser \\& Peebles (1973) has been modified and applied to the 2 year maps produced by the \\COBE\\ DMR.  The power spectrum of the real sky has been compared to the power spectra of a large number of simulated random skies produced with noise equal to the observed noise and primordial density fluctuation power spectra of power law form, with $P(k) \\propto k^n$. Within the limited range of spatial scales covered by the \\COBE\\ DMR, corresponding to spherical harmonic indices $3 \\leq \\ell \\lsim 30$, the best fitting value of the spectral index is $n = \\nTBDth$ with the Harrison-Zeldovich value $n = 1$ approximately 0.5$\\sigma$ below the best fit. For $3 \\leq \\ell \\lsim 19$, the best fit is $n = \\nTBD$. Comparing the \\COBE\\ DMR $\\Delta T/T$ at small $\\ell$ to the $\\Delta T/T$ at $\\ell \\approx 50$ from degree scale anisotropy experiments gives a smaller range of acceptable spectral indices which includes $n = 1$. ",
+        "introduction": "The spatial power spectrum of primordial density perturbations, $P(k)$ where $k$ is the spatial wavenumber, is a powerful tool in the analysis of the large scale structure in the Universe. In the first moments after the Big Bang, the horizon scale $ct$ corresponds to a current scale that is much smaller than galaxies, so the assumption of a scale free form for $P(k)$ is natural, which implies a power law $P(k) \\propto k^n$. Harrison (1970), Zeldovich (1972), and Peebles \\& Yu (1970) all pointed out that the absence of tiny black holes implies $n \\lsim 1$, while the large-scale homogeneity implied by the near isotropy of the Cosmic Microwave Background Radiation (CMBR) requires $n \\gsim 1$. Thus the prediction of a Harrison-Zeldovich or $n = 1$ form for $P(k)$ by an analysis that excludes all other possibilities is an old one. This particular scale-free power law is scale-invariant because the perturbations in the metric (or gravitational potential) are independent of the scale. The inflationary scenario of Guth (1981) proposes a tremendous expansion of the Universe (by a factor $\\geq 10^{30}$) during the inflationary epoch, which can convert quantum mechanical fluctuations on a microscopic scale during the inflationary epoch into Gpc-scale structure now.  To the extent that conditions were relatively stable during the small part of the inflationary epoch which produced the Mpc to Gpc structures we now study, an almost scale-invariant spectrum is produced (Bardeen, Steinhardt \\& Turner 1983). Bond \\& Efstathiou (1987) show that the expected variance of the coefficients $a_{\\ell m}$ in a spherical harmonic expansion of the CMBR temperature given a power law power spectrum $P(k) \\propto k^n$ is $<a_{\\ell m}^2> \\; \\propto \\Gamma[\\ell+(n-1)/2] / \\Gamma[\\ell+(5-n)/2]$ for $\\ell < 40$. Thus a study of the angular power spectrum of the CMBR can be used to place limits on the spectral index $n$ and test the inflationary prediction of a spectrum close to the Harrison-Zeldovich spectrum with $n = 1$.  The angular power spectrum contains the same information as the angular correlation function, but in a form that simplifies the visualization of fits for the spectral index $n$. Furthermore, the off-diagonal elements of the covariance matrix have a smaller effect for the power spectrum than for the correlation function. However, with partial sky coverage the multipole estimates in the power spectrum are correlated, and this covariance must be considered when analyzing either the correlation function or the power spectrum.  The power spectrum of a function mapped over the entire sphere can be derived easily from its expansion into spherical harmonics, but for a function known only over part of the sphere this procedure fails. Wright (1993) has modified a power spectral estimator from Peebles (1973) and Hauser \\& Peebles (1973) that allows for partial coverage and applied this estimator to the DMR maps of CMBR anisotropy. We report here on the application of these statistics to the DMR maps based on the first two years of data (Bennett \\etal\\ 1994). Monte Carlo runs have been used to calculate the mean and covariance of the power spectrum. Fits to estimate \\Amp\\ and $n$ by maximizing the Gaussian approximation to the likelihood of the angular power spectrum are discussed in this paper. Since we only consider power law power spectrum fits in this paper, we use $Q$ as a shorthand for \\Amp\\ or $Q_{rms-ps}$, which is the RMS quadrupole averaged over the whole Universe, based on a power law fit to many multipoles. \\Amp\\ should not be confused with the actual quadrupole of the high galactic latitude part of the sky observed from the Sun's location within the Universe, which is the $Q_{RMS}$ discussed by Bennett \\etal\\ (1992a). ",
+        "conclusions": "The Hauser-Peebles method of analyzing angular power spectra has been applied to the DMR maps. While the best fit to the observed power spectrum has $n = \\nTBDth$, this deviation from the $n = 1$ case is not statistically significant. To obtain a more accurate determination of $n$ we need to compare the \\COBE\\ DMR amplitude with ground-based and balloon-based experiments at smaller angular scales, which are sensitive to higher $\\ell$'s than \\COBE.  The ULISSE and Tenerife experiments (Watson \\etal\\ 1992) with beam sizes near $6^\\circ$ and chopper throws of $6-8^\\circ$ give upper limits in the $\\ell_{eff} \\approx 15$ range which support $n < 1.5$. The new Tenerife results (Hancock \\etal\\ 1994) at the same angular scale as Watson \\etal\\ give a central value that is slightly above the earlier upper limit.  Both are plotted in Figure \\ref{logngpower}.   The calculations of Kamionkowski \\& Spergel (1994) suggest that for open Universes with $\\Omega \\approx 0.1$ the power at low $\\ell$'s will be depressed relative to the $n = 1$ flat Universe prediction. This prediction is consistent with the data presented here, but fluctuations due to cosmic variance at low $\\ell$'s are as large as the difference between the open Universe model and the scale-invariant flat Universe $n=1$ model.  The string model prediction given by Bennett, Stebbins \\& Boucher (1992b) also has lower power at small $\\ell$'s, and is thus consistent with the \\COBE\\ angular power spectrum, but a cutoff at higher $\\ell$ is needed. Reionization of the Universe at redshift $z$ will hide structures on scales smaller than $60^\\circ/\\sqrt{1+z}$ and provide the needed cutoff, but $z \\gsim 100$ is required to have a substantial optical depth with the baryon abundance derived from Big Bang nucleosynthesis (Walker \\etal\\ 1991). But reionization will not ``smear out'' the edges produced by strings seen at smaller redshifts. Thus, unlike most scientific models which can only be falsified, the string model can be {\\em verified} by finding ``the edge'', which will remain infinitely sharp even with reionization. The sharp edges in the $\\Delta T$ maps produced by nearby strings limits the slope of the cutoff to $\\ell^{-1}$ relative to an $n = 1$ spectrum. Graham \\etal\\ (1994) find that the SP91 data is significantly non-Gaussian, which suggests that an edge may have been found.  If true, this would increase the discrepancy among the degree-scale experiments, since the presence of an edge would increase the variance, but SP91 has the smallest variance of the four degree-scale experiments.  This expected increase in the variance due to non-Gaussian features is clearly present in the 20 GHz OVRO and RING experiments which have the same angular scale.  The RING experiment covered a larger region, and was contaminated by discrete sources whose existence was verified by the VLA. The 170 GHz MSAM experiment also saw what appeared to be discrete sources, and these were not included in the analysis. There is no sensitive, higher angular resolution telescope to verify that the large deflections seen by MSAM are indeed point sources. Thus it is possible that the large deflections are true cosmic $\\Delta T$'s. The open circles above the MSAM data points in Figure \\ref{logngpower} show the increased power that results if these data are not excluded in the analysis.  The bulk flow data of Bertschinger \\etal\\ (1990) (at $\\ell_{eff} \\approx 10^2$) require $n \\approx 1$ to agree with \\COBE\\ (Wright \\etal\\ 1992), while the larger bulk flow on larger scales seen by Lauer \\& Postman (1992) requires $n \\approx 2.9$ to agree with \\COBE, if we assume that the reported bulk flow represents the RMS velocity on this scale.  The experiments at $\\approx 1^\\circ$  scale offer the possibility of a better determination of the primordial power spectrum index $n$, but the model-dependent effects of the wing of the Doppler peak at $\\ell \\approx 200$ must be allowed for. Even in the large angle region $\\ell < 30$ small model-dependent corrections must be made.  In Figure \\ref{logngpower}, the upper Cold Dark Matter (CDM) curve has a primordial spectral index $n_{pri} = 0.96$, but an apparent index $n_{app}$ = 1.1. Since the spectral index \\nTBDth\\ found in this paper is an apparent index, the \\COBE\\ power spectrum is consistent with the prediction from inflation and CDM or Mixed Dark Matter models that $n_{app} \\approx 1.1$. On the other hand, vacuum-dominated models such as the Holtzman (1989) model with $\\Omega_B = 0.02$, $h = 1$, $\\Omega_{CDM} = 0.18$, and $\\Omega_{vac} = 0.8$ will give $n_{app} = 0.9$ for $n_{pri} = 1.0$ (Kofman \\& Starobinski 1985), which deviates by slightly less than $1\\sigma$ from the \\COBE\\ value. We have found the values of the primordial spectral index $n_{pri}$ that will connect the \\COBE\\ NG amplitude and $\\ell_{eff}$ found earlier with the degree-scale experiments using the scalar transfer function from Crittenden \\etal\\ (1993). We have ignored the tensor transfer function because the current accuracy in determining $n$ is not sufficient to fix the tensor to scalar ratio, and because the excess quadrupole predicted by the tensor transfer function is not seen in the \\COBE\\ power spectrum. The South Pole experiment of Schuster \\etal\\ (1993) at $\\ell \\approx 44$ requires $n_{pri} \\approx 0.48 \\pm 0.34$ to agree with \\COBE. The Saskatoon experiment of Wollack \\etal\\ (1993) at $\\ell \\approx 55$ requires $n_{pri} \\approx 1.04 \\pm 0.29$ to agree with \\COBE. The PYTHON experiment of Dragovan \\etal\\ (1994) at $\\ell \\approx 71$ requires $n_{pri} \\approx 1.58 \\pm 0.22$ to match \\COBE. The ARGO experiment of de Bernardis \\etal\\ (1994) at $\\ell \\approx 75$ requires $n_{pri} \\approx 1.10 \\pm 0.16$ to match \\COBE. The weighted mean of these values is $n_{pri} = 1.15 \\pm 0.11$. Unfortunately $\\chi^2 = 8.0$ with 3 degrees of freedom when comparing these four values of $n_{pri}$ with this weighted mean, indicating that these four experiments are mutually inconsistent. If we allow for this discrepancy by scaling the error on $n_{pri}$, we get a value $n_{pri} = 1.15 \\pm 0.2 \\pm 0.1$ from this comparison of \\COBE\\ with the degree-scale experiments, where the second error bar is contribution of the uncertainty of the \\COBE\\ NG amplitude to $n_{pri}$. Thus this comparison of \\COBE\\ with degree-scale experiments gives a more precise value the primordial spectral index that is still consistent with inflation. With more data from COBE (4 years are recorded) the large angular scale amplitude will become more and more certain.  The $\\ell_{eff}$ associated with this amplitude will shift to larger values $\\approx 10$. Reliable, consistent determinations of $\\delta T$ on scales $\\ell_{eff} \\approx 50$ will be needed to compare with the large-scale $\\Delta T$. With only two years of data, the \\COBE\\ DMR large scale amplitude has relative errors that are two times smaller than the errors of the current degree-scale experiments. Thus the degree-scale experiments need to be extended to a sky coverage that is 10 times higher than their current coverage to match the expected \\COBE\\ uncertainty with four years of data, or else achieve an equivalent increase in accuracy by reduced noise or systematic errors.  \\bigskip  We gratefully acknowledge the many people who made this paper possible: the NASA Office of Space Sciences, the {\\it COBE} flight operations team, and all of those who helped process and analyze the data. In particular we thank Tony Banday, Krys G\\'orski, Gary Hinshaw, Charlie Lineweaver, Mike Hauser, Mike Janssen, Steve Meyer and Rai Weiss for useful comments on the manuscript.  \\clearpage  \\begin{table*} \\plotone{table.ps} \\caption{} \\label{bigtab} \\end{table*}  \\clearpage  \\begin{table*} \\begin{center} \\begin{tabular}{ccrlrlrlrlrl} $\\ell$ & $\\ell_{eff}$ & \\multicolumn{2}{c}{2 YR 53} & \\multicolumn{2}{c}{2 YR 53+90} & \\multicolumn{2}{c}{2 YR 53$\\times$90} & \\multicolumn{2}{c}{2 YR NG} & \\multicolumn{2}{c}{1 YR 53+90}\\\\ \\tableline 2    & 2.1  & 0.59 & $\\pm 0.47$ & 0.44 & $\\pm 0.61$ & 0.45  & $\\pm 0.58$ & 0.17  & $\\pm 0.49$ & 0.58 & $\\pm 0.42$ \\\\ 3    & 3.1  & 1.06 & $\\pm 0.52$ & 1.04 & $\\pm 0.60$ & 1.01  & $\\pm 0.56$ & 0.96  & $\\pm 0.56$ & 0.90 & $\\pm 0.50$ \\\\ 4    & 3.3  & 1.16 & $\\pm 0.45$ & 1.11 & $\\pm 0.52$ & 1.12  & $\\pm 0.48$ & 1.05  & $\\pm 0.52$ & 1.14 & $\\pm 0.46$ \\\\ 5-6   & 4.4  & 1.23 & $\\pm 0.40$ & 1.21 & $\\pm 0.40$ & 1.22  & $\\pm 0.36$ & 1.00  & $\\pm 0.54$ & 1.06 & $\\pm 0.42$ \\\\ 7-9   & 6.2  & 1.03 & $\\pm 0.43$ & 1.23 & $\\pm 0.37$ & 1.19  & $\\pm 0.33$ & 1.26  & $\\pm 0.63$ & 1.19 & $\\pm 0.46$ \\\\ 10-13 & 8.7  & 1.31 & $\\pm 0.58$ & 1.51 & $\\pm 0.44$ & 1.15  & $\\pm 0.41$ & -0.22 & $\\pm 1.12$ & 1.41 & $\\pm 0.71$ \\\\ 14-19 & 10.7 & 2.61 & $\\pm 0.91$ & 1.60 & $\\pm 0.67$ & 1.69  & $\\pm 0.65$ & 1.90  & $\\pm 2.34$ & 2.34 & $\\pm 1.27$ \\\\ 20-30 & 11.8 & 3.26 & $\\pm 2.60$ & -1.14 & $\\pm 1.98$ & 0.58  & $\\pm 1.94$ & -2.54  & $\\pm 7.83$ & -0.19 & $\\pm 3.98$ \\\\ \\hline \\end{tabular} \\end{center} \\caption{Ratio of the binned power spectrum from Equation 9 to a $Q = 17\\;\\muK$, $n = 1$ model.} \\label{pstable} \\end{table*}  \\begin{table*} \\begin{center} \\begin{tabular}{rrrrrrrrr} $\\ell=$ & 2 & 3 & 4 & 5-6 & 7-9 & 10-13 & 14-19 & 20-30 \\\\ \\tableline 2 yr NG  &    55 &   208 &   234 &   319 &   365 &   -45 &   251 &  -161 \\\\ Best fit &   202 &   159 &   170 &   272 &   286 &   229 &   167 &    85 \\\\ Q=17,n=1 &   316 &   216 &   225 &   318 &   290 &   202 &   132 &    63 \\\\ \\\\ Noise only &  1374 &     1 &   253 &   181 &   -42 &   -28 &   -12 &    -3 \\\\ &       &  1660 &    11 &   745 &   213 &    -1 &   238 &   311 \\\\ Signal \\& Noise & 24871 &       &  2164 &   669 &    30 &   133 &   171 &  -186 \\\\ &   364 & 14803 &       & 10865 &  2018 &   139 &   226 & -1782 \\\\ &  3611 &  -166 & 13599 &       & 18437 &  2307 &   -28 & -1230 \\\\ &  1717 &  3922 &  1214 & 29146 &       & 42218 &  5603 &   264 \\\\ &   828 &  1426 &   720 &  3907 & 33123 &       & 87028 & 10292 \\\\ &  2055 &  -187 &   384 &  1078 &  3798 & 53633 &       & 241805 \\\\ &  1128 &   776 &  -354 &   949 &  -241 &  7243 & 95862 &       \\\\ & -351  &  -211 &   430 & -2326 & -1799 &   229 &  9420 & 245629 \\\\ \\hline \\end{tabular} \\end{center} \\caption{Binned Hauser-Peebles power spectrum of the 2 year NG maps, the best fit $n=1.4$ model, the nominal $Q=17, n=1$ model, all in $\\muK^2$; the upper triangle of the covariance matrix from noise-only Monte Carlo runs, and the lower triangle of the covariance matrix from the best fit Monte Carlo runs in $\\muK^4$.} \\label{cvrtable} \\end{table*}  \\begin{table*} \\begin{center} \\begin{tabular}{llcll} Method & COBE dataset & Q? & Result & Reference \\\\ \\tableline Correlation function & 1 year 53$\\times$90 & N & $n_{app} = 1.15^{+0.45}_{-0.65}$ & Smoot \\etal\\ (1992) \\\\ COBE:$\\sigma_8$      & 1 year 53+90        & N & $n_{pri} = 1 \\pm 0.23 $ & Wright \\etal\\ (1992) \\\\ Genus \\vs\\ smoothing & 1 year 53           & Y & $n_{app} = 1.7^{+1.3}_{-1.1}$ & Smoot \\etal\\ (1994)  \\\\ RMS \\vs\\ smoothing   & 1 year 53           & Y & $n_{app} = 1.7^{+0.3}_{-0.6}$ & Smoot \\etal\\ (1994)  \\\\ Correlation function & 2 year 53$\\times$90 & Y & $n_{app} = 1.59^{+0.49}_{-0.55}$ & Bennett \\etal\\ (1994) \\\\ Correlation function & 2 year 53$\\times$90 & N & $n_{app} = 1.21^{+0.60}_{-0.55}$ & Bennett \\etal\\ (1994) \\\\ COBE\\,:\\,$1^\\circ$ scale & 2 year NG           & N & $n_{pri} = 1.15\\pm 0.2$ & this paper \\\\ Cross power spectrum & 1 year (53A+90A)$\\times$(53B+90B) & N & $n_{app} = 1.69^{+0.45}_{-0.52} $  & this paper \\\\ Cross power spectrum & 2 year 53A$\\times$53B & N & $n_{app} = 1.41^{+0.75}_{-1.17} $  & this paper \\\\ Cross power spectrum & 2 year 53$\\times$90 & N & $n_{app} = 1.22^{+0.42}_{-0.46} $  & this paper \\\\ Cross power spectrum & 2 year (53A+90A)$\\times$(53B+90B) & N & $n_{app} = 1.32^{+0.39}_{-0.45} $  & this paper \\\\ Cross power spectrum & 2 year (53A+90B)$\\times$(53B+90A) & N & $n_{app} = 1.20^{+0.42}_{-0.46} $  & this paper \\\\ Cross power spectrum & 2 year NGA$\\times$NGB & N & $n_{app} = 1.41^{+0.75}_{-1.17} $  & this paper \\\\ Orthonormal functions & 2 year 53+90 & N & $n_{app} = 1.02 \\pm 0.4 $ & G\\'orski \\etal\\ (1994) \\\\ \\hline \\end{tabular} \\end{center} \\caption{Spectral index determinations} \\label{ntable} \\end{table*}  \\clearpage"
+    },
+    "9401/astro-ph9401034_arXiv.txt": {
+        "abstract": "We describe a refined calculation of high energy emission from rotation-powered pulsars based on the Outer Gap model of Cheng, Ho \\&~Ruderman (1986a,b). In this calculation, vacuum gaps form in regions near the speed-of-light cylinder of the pulsar magnetosphere along the boundary between the closed and open field line zones. We have improved upon previous efforts to model the spectra from these pulsars (e. g. Cheng, et al. 1986b; Ho 1989) by following the variation in particle production and radiation properties with position in the outer gap. Curvature, synchrotron and inverse-Compton scattering fluxes vary significantly over the gap and their interactions {\\it via} photon-photon pair production build up the radiating charge populations at varying rates. We have also incorporated an approximate treatment of the transport of particle and photon fluxes between gap emission zones. These effects, along with improved computations of the particle and photon distributions, provide very important modifications of the model gamma-ray flux. In particular, we attempt to make specific predictions of pulse profile shapes and spectral variations as a function of pulse phase and suggest further extensions to the model which may provide accurate computations of the observed high energy emissions. ",
+        "introduction": "Since the discovery of radio pulsations from rotating neutron stars (Hewish et al.~1968), many theories have been proposed to explain the origin of these emissions. Early models invoke the acceleration of electrons and positrons to extremely high energies such that the radiation emitted by these particles results in an electromagnetic cascade of $e^\\pm$ pairs and photons. This avalanche of charges yields the observed coherent radio flux (Sturrock 1971; Ruderman \\& Sutherland 1975). Unfortunately, attempts at understanding the physics of the pulsar magnetosphere by looking only at the radio observations have met with limited success.  A more fruitful approach would be to use the higher energy observations, both in terms of photon energy and in terms of the total available energy to be extracted from the pulsar, to characterize the magnetosphere (Arons 1992).  The polar cap models of Harding, Tademaru, \\&~Esposito (1978), Daugherty \\&~Harding (1982), the slot gap model of Arons and his collaborators (Scharlemann, Arons, \\&~Fawley 1978; Arons 1983) and the outer gap model of Cheng, et al. (1986a,b; hereafter CHRa,b) and Ho (1989) have all been attempts to use the high energy observations to reveal nature of the pulsar magnetosphere.  Given the recent data obtained by the instruments aboard the {\\it Compton Gamma-Ray Observatory} and the identification of additional gamma-ray pulsars, bringing the total to six---Crab, Vela, Geminga (Bertsch et al. 1992), PSRB~1706$-$44 (Thompson et al. 1992), PSRB~1055$-$52 (Fierro et al. 1993) and PSRB~1509$-$58 (Wilson et al. 1992)---it is now appropriate for further efforts to be made towards understanding these objects.  In this paper, we present our own attempts at modeling the high energy emission from young, rotation-powered pulsars in the context of a modified outer gap model. In section~2, we summarize the results of a previous paper (Chiang \\& Romani~1992; hereafter Paper~I), in which a geometrical calculation of the emission from the outer gap regions was performed and which was shown to reproduce qualitatively the light curves of all the gamma-ray pulsars. In section~3, we review the details of the outer gap spectral calculation described in CHRa,b and restated later by Ho (1989); and in section~4, we motivate and outline our refinements to the CHR calculation. They consist of dividing the outer gap region into smaller sub-zones in order to account for the variation in magnetic field and photon and particle densities as a function of position in the magnetosphere. We have also included a treatment of the transport of particles and photons from one region of the magnetosphere to another.  We find that this transport crucially effects the emission processes in the outer gap.  In addition, we perform more detailed radiation and pair creation calculations. In section~5, we present some preliminary results of our multi-zone spectral calculations; and in section~6, we discuss limitations of the model and ways in which it can be further improved. ",
+        "conclusions": "We have described our efforts to refine the standard outer gap picture of high energy emission from pulsar magnetospheres. Our light curve calculations, in our more general outer gap geometry, are able to account for the various pulsar profiles seen for the gamma-ray pulsars observed by the instruments aboard the {\\em Compton Observatory}. These light curve calculations also point to the source of the observed spectral variation with phase seen for the gamma-ray pulsars: there is a clear mapping of location in the magnetosphere to pulse phase. We have thus attempted to carry through the outer gap spectral calculation as outlined by CHRb and Ho (1989) in our modified geometry, including the refinements of radiation and particle transport and improved inverse-Compton and pair-production calculations. Our efforts have met with limited success. The calculation suffers from too little low energy photon flux and the converged pulsar spectra we produce are too hard for all phases of the pulse profile.  However, our results indicate that a realistic computation of the pulsar emission requires at least an approximate three-dimensional treatment of the radiation transport in the outer magnetosphere. Since photon fluxes themselves provide the relevant opacities, this is a computationally difficult, non-linear problem. However, the spectral differences between the competing radiation processes illustrated in our sample calculations do point the way for a more satisfactory description of the high-quality pulsar data from the {\\em Compton Gamma-Ray Observatory}. \\bigskip \\bigskip  We thank Cheng Ho for participating in important initial discussions on the physics of gap zones and for sharing a copy of his outer gap spectral code. RWR was supported in part by NASA grants NAGW-2963 and NAG5-2037. JC was supported by NASA grant NAG5-1605 and gratefully acknowledges useful discussions with {\\em EGRET} team members at Stanford University, Goddard Space Flight Center and the Max-Planck-Institut f\\\"ur Extraterrestriche Physik at Garching, Germany.   \\clearpage"
+    },
+    "9401/astro-ph9401020_arXiv.txt": {
+        "abstract": " ",
+        "introduction": "Two independent groups have recently found a total of three candidate microlensing events, apparently caused by Massive Compact Halo Objects (\\Ms) along the line of sight toward the Large Magellanic Cloud (LMC) (Alcock et al.\\ 1993; Aubourg et al.\\ 1993).  The events are achromatic, have maximum magnifications of $A_\\max = 6.8$, 3.3, and 2.5, and characteristic times $\\omega^{-1} = 17$ days, 30 days and 26 days. The light curves fit the  form first predicted by Paczy\\'nski (1986) quite well: \\begin{equation} A[x(t)] = {x^2 + 2\\over x(x^2 + 4)^{1/2}},\\qquad x(t) = \\sqrt{\\beta^2 + \\omega^2(t-t_0)^2}. \\end{equation} Here $t_0$ is the midpoint of the event, $\\omega^{-1}$ is the characteristic time, and $\\beta$ is the dimensionless impact parameter (normalized to the Einstein ring radius).  Of the three measurable parameters in equation (1), only the characteristic time, $\\omega^{-1}$ yields any information about the \\Msk. It is related to the underlying physical parameters by \\begin{equation} \\omega^{-1} = {\\theta_*\\dol\\over v}= {[4GM \\dol(1-\\dol/\\dlmc)]^{1/2}\\over v c}, \\end{equation} where $\\theta_*$ is the Einstein ring radius, $M$ is the \\M mass, $\\bf v$ is its transverse velocity, and $\\dol$ is its distance from the observer.  The distance of the lensed star is denoted $\\dlmc$.  For any given event, one cannot determine the mass, distance, and speed separately.  For an ensemble of events, the typical speed and distance are expected to be $v\\sim 200\\, \\kms$ and $\\dol\\sim 10$ kpc, respectively.  Hence, $M\\sim (70 \\omega\\, {\\rm days})^{-2} M_\\odot$. If the first few events prove typical, it will imply a mass scale of $\\sim 0.1\\, M_\\odot$, near the hydrogen-burning limit.  One would like to determine as much as possible about the distribution of \\Msk, not just the mass scale.  For example, one would like to know if the \\Ms  actually lie in a halo as opposed to a disk or thick disk, whether the halo is spherical or oblate, whether the  halo is truncated or extends all the way to the LMC, whether it is rotating, whether \\Ms populate the LMC as well as the Galaxy, and what the distribution of \\M masses is. To extract these additional pieces of information, new methods of analyzing the data as well as new experiments are required.  Gould  (1992, 1993, 1994a,b), Sackett \\& Gould (1993), and Gould, Miralda-Escud\\'e \\& Bahcall (1994) have developed a number of different techniques for extracting additional parameters from lensing events. In particular, Gould (1994a) showed that the non-zero size of the lensed stellar disk modifies the light curve of the event in a manner that allows one to measure the \\Mk's Einstein radius, $\\theta_*$, and hence its proper motion (angular speed), $\\omega\\theta_*$. In practice, however, the effect is measurable only for a fraction of events during  which the \\M passes  over the line of sight to the face of the source star, $(1/550)(M/0.1\\,M_\\odot)^{-1/2}$ for \\Ms in the Galactic halo and $(1/70)(M/0.1\\,M_\\odot)^{-1/2}$ for \\Ms in the LMC.  In this {\\it Letter}, we predict a second, spectroscopic, effect  which also arises from the non-zero size of the source, combined with the source rotation. Differential magnification of the stellar disk during the event  induces a shift in the stellar spectral lines.  By measuring the shift as a function of time one can determine the Einstein ring radius and so the proper motion of the \\Mk. In contrast to the photometric effect analyzed by Gould (1994a), this spectroscopic effect can be measured even when the \\M is many stellar radii from the source.  Hence, the proper-motion measurement can be made for a larger fraction of events. We estimate that with  an 8m telescope the effect can be measured in $\\sim 7\\%$ of photometrically detected relatively high-magnification  ($\\beta \\ltorder 0.5$) \\M events. The effect can be measured in $\\sim 15\\%$ of high-magnification events  generated by \\Ms in the LMC.  Measurement of the proper motion of Galactic \\Ms would, in itself, remove some of the degeneracy among the three \\M parameters, mass, distance, and transverse speed.  However, if this measurement were combined with a parallax measurement of the ``reduced velocity'', then the degeneracy could be completely broken and the three parameters plus the transverse direction could be separately determined.  The proper motion of Galactic \\Ms is $\\sim 15$ times greater than that of LMC \\Msk.  Hence, a proper-motion measurement clearly distinguishes between the two. By identifying even a handful of LMC \\Msk, one could determine the fraction of events generated by them.  This would both provide direct information about the LMC halo and remove a major background to the primary Galactic signal (see Sackett \\& Gould 1993). ",
+        "conclusions": "We predict the existence of a spectroscopic line-shift, resulting from differential magnification of a rotating stellar disk, during microlensing events in the ongoing MACHO and EROS experiments. The effect can be measured and monitored on large telescopes for a fraction of the A-star lensing events (A-stars should supply $\\sim 1/3$ of the events), provided that the spectroscopic observations can be initiated fast enough to catch the event before maximum light. The spectroscopic signature can serve as an additional test of the lensing (as opposed to variable-star) nature of the event. More importantly, the Einstein ring  radius can be measured from the temporal behavior of the line-shift. This technique can therefore remove part of the degeneracy in the derivation of \\M parameters in the ongoing experiments.  The spectral shift can be measured with relatively smaller effort (one measurement per night) in a  large fraction of microlensing events involving \\Ms in the halo of the LMC itself, which can then be recognized as such. If, for example, 10\\% of the events are due to LMC \\Msk, spectroscopic measurements of 50 A-star events with  $\\beta < 0.2$ for one hour per night for a week will reveal the effect in  $\\sim 5$ cases, and would provide a direct measurement of the LMC fraction of \\M events.  The LMC events can be distinguished from any Galactic ``foreground'' events by their much lower proper motion (Gould 1994a).  With a more ambitious program in which events are followed for a week with eight measurements per night (likely requiring an 8m telescope), the proper motion can also be measured for the A-star + Galactic \\M events with $\\beta\\ltorder 0.10$. While this is a lot of telescope time the potential payoff is very great, particularly if the ``reduced velocity'' can be measured from parallax observations (Gould 1992, 1994b; Gould et al.\\ 1994).  The reduced speed is given by $(D_{\\rm OL}^{-1}-D_{\\rm LMC}^{-1})^{-1} \\omega\\theta_*$.  Hence, by measuring the proper motion, $\\omega\\theta_*$ one would find the distance to the \\Mk, $\\dol$.  The distance and the Einstein radius, $\\theta_*$, yield the mass, $M$.  Combining these with $\\omega$ gives the transverse speed.  In general, measurement of the reduced velocity requires observation from two small satellites in solar orbit.  It is likely that the reduced speed could be obtained from only one such satellite (Gould 1994b).  Gould et al.\\ (1994) show that it is generally possible to measure one component of the reduced velocity from ground-based observations provided that the \\Ms are going of order the Earth's orbital speed of $30\\,\\rm km\\,s^{-1}$.  This would be extremely rare for \\Ms in the halo.  However, the measurement proposed by Gould et al.\\ actually becomes much more sensitive for very high-magnification events  $(\\beta\\ltorder 0.1)$ such as those for which proper-motion determinations can be made.  For these events, we conjecture that one or perhaps even both components of the transverse velocity could be measured from the ground, even for \\Ms traveling at halo speeds, $\\sim 200\\,\\rm km\\,s^{-1}$.  {\\bf Acknowledgements}: We would like to thank D.\\ Goldberg, T.\\ Mazeh, and C.\\ Pryor for helpful discussions.  D.M. acknowledges support through an Alon Fellowship.  \\newpage"
+    },
+    "9401/astro-ph9401013_arXiv.txt": {
+        "abstract": "We present a new analysis of previously published optical and radio data sets of the gravitationally lensed quasar 0957+561 A,B with the aim of determining the time delay between its two images. We use a non-parametric estimate of the dispersion of the combined data set where, however, we only make use of alternating neighbours in order to avoid windowing effects. From the optical data a time delay of (415 $\\pm$ 32) days and from the radio data a delay of (409 $\\pm$ 23) days is suggested. We demonstrate a considerable sensitivity of different delay estimation procedures against the removal of only a few observational data points or against smoothing or detrending of the original data sets. The radio data give us formally a slightly more precise value for the time delay than the optical data. Also, due to the lack of windowing effects, the result obtained for the radio data can be considered as somewhat more reliable than the delay determined from the optical data. ",
+        "introduction": "The time delay between the images of a gravitationally lensed object is of great astrophysical interest since it may be used to determine the Hubble parameter as well as the mass of the lens (Refsdal \\cite{Refsdala}, \\cite{Refsdalb}, Borgeest \\cite{Borgeest}, Falco et al.~\\cite{Falco91b}). The double quasar 0957+561 A,B (Walsh et al.~\\cite{Walsh}, Young \\cite{Young}, Falco \\cite{Falco92}) is up to now the only gravitational lens system for which serious attempts have been made to determine the time delay $\\tau$ between its images. However, the results are still controversial, with suggested time delays of between 376 and 657 days (Florentin-Nielsen \\cite{Florentin}, Schild \\& Cholfin \\cite{Schild86}, Gondhalekar \\cite{Gondh}, Gorenstein et al. \\cite{Goren}, Leh\\'ar et al. \\cite{Lehar89}, Vanderriest \\cite{Van89}, Schild \\cite{Schild90}, Falco et al. \\cite{Falco91}, Leh\\'ar et al. \\cite{Lehar92}, Roberts et al. \\cite{Roberts}, Press et al. \\cite{Pressa}, \\cite{Pressb}, Beskin \\& Oknyanskij \\cite{Beskin}). Using an elaborate statistical method (see Rybicki et al. \\cite{Rybicki}) Press et al. (\\cite{Pressa}, below PRHa), recently obtained a value of ($536\\pm 12$) days from the optical lightcurve, which is seemingly in good agreement with the value obtained from the radio data by the same authors (Press et al. \\cite{Pressb}, below PRHb). They conclude that ``delays less than about 475 days are strongly excluded''. It is also often claimed (Leh\\'ar \\cite{Lehar92}, Roberts et al. \\cite{Roberts}, PRHb) that the radio data decisively exclude a delay of around 415 days, a value favoured by other authors (Vanderriest et al. \\cite{Van89}, Schild 1990).  The time delay obtained by PRHa is quite close to 1.5 years, a value for which windowing effects due to the uneven sampling of the optical data are expected to be strongest (Vanderriest et al. \\cite{Van92}).  We here present results of a careful re-analysis of the same data sets as used in PRHa and PRHb using simple explorative type statistical methods. The main emphasis of our work is the evaluation of two competing hypothetical time delays: 415 and 536 days.  For the combined data set generated from the data of image A and the data of image B, time shifted by $\\tau$, we basically estimate the dispersion $D^2$ of the scatter around the unknown mean curve. The true time delay between the images should then show up as a minimum in the dispersion spectrum $D^2(\\tau)$. It is our aim to determine the dispersion of the combined data due to the alternation between the two light curves, and not the dispersion within each lightcurve. In order to avoid strong windowing effects we therefore take into account only alternating neighbouring pairs in the combined data set, i.e. only pairs where one point is from A and the other one from B, respectively. It should be noted that for $\\tau=(n+0.5)$ years (where $n$ is integer) the dispersion $D^2(\\tau)$ would otherwise be strongly dominated by the inner dispersion of the two original light curves, naturally leading to pronounced minima.  In Section 2 of the paper we introduce the basic statistical techniques used. In Section 3 we present the results of a detailed analysis of the light curves of the double quasar 0957+561 A,B. We here use the same data sets as PRHa and PRHb to allow a comparison of the results and to understand the problems leading to the different results. In the concluding part we try to summarize the work done on time delay estimation up to now and to delineate some inner difficulties of the problem under study. Technical details of the nonparametric dispersion estimation are treated in the Appendix. ",
+        "conclusions": "We introduced an extremely simple, but nevertheless sensitive method to seek probable time delay values for light curves which are assumed to originate from one and the same source. The method easily recovered results which were obtained by more complex and statistically sound procedures presented in PRHa and PRHb. However, the simplicity of the proposed method allowed us to get additional insights into the problems which originate from unfortunate spacing of the sampling points for the optical light curves. It allowed us also to produce an alternative consistent solution for the time delay problem.  The dispersion spectra of the optical and radio data can show minima in various places depending on the method of analysis and preprocessing of the data. For theoretical reasons delays $\\tau \\leq 0$ can be excluded. The minimum near 1.5 years most probably results from a statistical fluctuation which is amplified due to the windowing, since for this time shift the overlap of data from the two light curves is at a minimum and thus the least squares minimization has more freedom to fit free parameters.  For the radio data we found that removing only two observations from B gives a stable spectrum for the dispersions, with a best value for the shift near $\\tau=409$ days. This result is in good agreement with the value of $\\tau=415$ days for the optical curve, which reveals itself after the detrending of the original data, or in other circumstances which allow to depress the instable minimum around $536$ days.  We think that the time delay controversy on QSO 0957+561 A,B is still not settled, but there is quite strong evidence (especially if we take into account preliminary results obtained using the extended data set by R. Schild) that the time delay is in the region of 400-420 days."
+    },
+    "9401/astro-ph9401045_arXiv.txt": {
+        "abstract": "There are now a half dozen young pulsars detected in high energy photons by the Compton GRO, showing a variety of emission efficiencies and pulse profiles.  We present here a calculation of the pattern of high energy emission on the sky in a model which posits $\\gamma$-ray production by charge depleted gaps in the outer magnetosphere. This model accounts for the radio to $\\gamma$-ray pulse offsets of the known pulsars, as well as the shape of the high energy pulse profiles. We also show that $\\sim 1/3$ of emitting young radio pulsars will not be detected due to beaming effects, while $\\sim 2.5 \\times$ the number of radio-selected $\\gamma$-ray pulsars will be viewed only high energies.  Finally we compute the polarization angle variation and find that the previously misunderstood optical polarization sweep of the Crab pulsar arises naturally in this picture. These results strongly support an outer-magnetosphere location for the $\\gamma-$ray emission. ",
+        "introduction": "Since the birth of $\\gamma$-ray astronomy the two most prominent galactic point sources have been identified with young rotation-powered pulsars, the Crab and Vela pulsars. The site of $\\gamma$-ray production has however been intensely debated for almost two decades. The two principal models are the `polar cap' picture exemplified by the work of Daugherty and Harding (1982) and the `outer gap' model championed by Cheng, Ho and Ruderman (1988, CHR). In the former acceleration occurs near the neutron star surface and the observed emission results from a $\\gamma-B$ pair cascade. In the second model acceleration is posited in charge depleted regions near the speed of light cylinder and $\\gamma-\\gamma$ pair production is an important process. We have examined these two models (Chiang and Romani 1992) and find that the observed pulse profiles arise most naturally in a modified version of the outer gap picture. Moreover, spectral calculations (Chiang and Romani 1994, CR94) have shown that in this model emission processes vary throughout the magnetosphere and can produce spectral variations through the pulse like those seen. In this paper we quantitatively compare the outer gap predictions with the observed profiles, supporting these conclusions.  The impetus for these computations is the dramatic increase in information on the $\\gamma$-ray pulsars provided by the {\\it Compton Gamma Ray Observatory}. In addition to improved light curves and phase resolved spectra for Crab and Vela ({\\it e.g.} Nolan \\et  1993), CGRO has detected at least four other pulsars in high-energy emission, including Geminga, which has not been seen in radio emission ({\\it e.g.} Thompson, \\et 1992, Ulmer, \\et 1993, Fierro, \\et 1993, Mayer-Hasselwander \\et 1994). Other bright plane sources await confirmation as pulsars, and the expected number of each of these classes of objects is presently unknown. It is apparent from these data that older rotation-powered pulsars are increasingly efficient producers of $\\gamma$ emission, at least for characteristic ages $\\la 10^{5.5} - 10^6$y. In deciphering the puzzle of $\\gamma$ production efficiencies upper limits on other pulsars are also important. Their interpretation is, however, unclear. Finally detailed information on the location of the emission region is an essential tool for understanding the polarization and spectrum seen from the observed pulsars. The geometrical computations of pulsar beaming in this paper address each of these issues. ",
+        "conclusions": "We have computed the expected pulse shapes and beaming fractions for $\\gamma$-ray emission from young radio pulsars in a model of charge acceleration in the outer magnetosphere. The results bear a strong resemblance to $\\gamma$-ray pulse profiles detected by CGRO. In addition our model reproduces the phase relationships between the radio emission and the $\\gamma$-ray pulse. Observed X-ray and optical pulsations find a natural interpretation in our geometric picture, as well. Important connections between radio pulse and polarization properties resulting from the viewing angle and expected $\\gamma$-ray profiles have also been established -- these give useful predictions for individual pulsars and allow strong tests of the model. We further find that the polarization properties of the Crab pulsar, with emission arising in the upper magnetosphere are easily explained in our model; this supercedes previously confusing interpretations of the polarization data. Finally, the numbers of objects detected in the radio and $\\gamma$-ray channels are as expected for a population of young pulsars and we infer that most of the unidentified galactic plane sources will be pulsars.  These results are based principally on the {\\it geometry} of the emission region and are thus relatively free from uncertainties in details of the emission process.  We are in the process of generating quantitative models for the radiation produced in the upper magnetosphere, but complicated non-linear models will be needed to give accurate results. Previous estimates of outer gap spectra and fluxes (CHR88, Ho 1990) give rough estimates of the total energy available, but do not give accurate results for the observable flux or spectrum.  Nonetheless, the dominant physical processes in this outer gap have already been identified by CHR88. The substantial amendments needed for a realistic model, however, mean that spectral results are not yet available.  We feel that the success of our geometrical sums firmly establishes the location of the $\\gamma$-ray production in the outer magnetosphere and thus resolves the long standing debate with polar cap models in favor of an outer gap picture.  Unless similar results can be duplicated by polar models, we feel that these models are not viable and efforts to compute the spectrum and luminosity from the polar cap site cannot explain the bulk of the observed pulsar emission. Much work remains to be done to provide a full picture of the origin of pulsar gamma rays, but assignment of the radiation site gives good hope for further progress. In particular CR94 show that the radiation processes vary strongly with altitude, and that these variations can be mapped directly to pulse phase. Thus phase resolved spectra provide a keen diagnostic for refining global emission models. Ultimately, detailed comparison with observed profiles should allow us to probe the inertia and current perturbations that are not followed in the present calculation. Understanding of these will help greatly in unraveling the mechanics of the pulsar acceleration process and in producing a self-consistent model of the pulsar magnetosphere. Thus $\\gamma$-ray measurements can become an important tool in deciphering the puzzle of the pulsar phenomenon."
+    },
+    "9401/astro-ph9401002_arXiv.txt": {
+        "abstract": "The thermal nucleation of quark matter bubbles inside neutron stars is examined for various temperatures which the star may realistically encounter during its lifetime. It is found that for a bag constant less than a critical value, a very large part of the star will be converted into the quark phase within a fraction of a second. Depending on the equation of state for neutron star matter and strange quark matter, all or some of the outer parts of the star may subsequently be converted by a slower burning or a detonation. ",
+        "introduction": "\\label{secintro}  If pulsars or the central parts of these can be made of quark matter rather than neutrons \\cite{witten84}, does this then apply to all or just some of them, and when and how does the phase transformation take place?  According to some investigations \\cite{horvath}, the transformation occurs during the supernova explosion. In this scenario, the released binding energy is what makes the supernova succeed in the first place, supplying the final ``push'' which seems to lack in most of the computer simulations of the events.  Another model for strange star formation (in the context of absolutely stable strange quark matter) was introduced by Baym {\\it et al.} \\cite{olinto}, describing the transformation as a slow burning (combustion) rather than a violent event connected with a supernova detonation.  Regardless of the way in which the transformation occurs, an initial seed of quark matter is needed to start it. Alcock {\\it et al.} \\cite{alcock} suggested a variety of possibilities ranging from pressure induced conversion via two flavor quark matter to collision with either highly energetic neutrinos or smaller lumps of strange quark matter. However, they did not provide a rate for conversion of neutron stars and thus left it as an open question, whether every compact object is a strange star, or whether they are rare objects, even if quark matter formation is energetically favorable. It has also been suggested \\cite{madsenfriedman} that strange matter seeds (in the case of quark matter stability) from strange star collisions or of cosmological origin would trigger the transformation of all neutron stars, in which case the thermal nucleation would be of relevance to the case of unstable quark matter only \\cite{hoenenogaegget}.  Other possibilities are that shock waves in the supernova trigger the conversion; a seed could be produced by non-thermal quantum fluctuations, or a phase transition could be started around impurities. We are not able to estimate the probability of either method, but would expect at least quantum fluctuations to be less likely than the thermal nucleation process discussed below.  An estimate for quark matter formed via thermally induced fluctuations was given by Horvath {\\it et al.} \\cite{deltamu}, using typical numbers for various physical quantities. It was found that all neutron stars are converted into strange stars (assuming stable strange quark matter) if the temperature at some time during the stars lifetime has exceeded 2-3 MeV.  In the following, we choose an approach similar to the one in Ref.\\ \\cite{deltamu}, but with an extra term in the expression for the surface energy of quark matter, and with the two phases treated in a more self-consistent way. Unlike most of the approaches mentioned above \\cite{witten84,horvath,olinto,alcock,madsenfriedman,deltamu}, we will be considering the formation of both strange stars (for absolutely stable strange matter) and hybrid stars, where strange matter is formed only in the central regions due to the high pressure. First, we will deal with some general aspects of nucleation (Sec.\\ \\ref{secboil}). For pedagogical purposes, Sec.\\ \\ref{pure} treats the problem using a simplified model with the hadron phase being a free degenerate neutron gas and the quark phase a bag model with only $u$ and $d$ quarks. A more detailed model for the neutron star is presented in Sec.\\ \\ref{seccompli}, followed by some concluding remarks (Sec.\\ \\ref{secclu}). ",
+        "conclusions": "\\label{secclu}  What we have seen is that if the bag constant lies in the interval where three flavor but not two flavor quark matter is stable at zero pressure and temperature (145 MeV $\\leq B^{1/4} \\leq$ 163 MeV, see Ref.\\ \\cite{farhijaffe}) then all or parts of a neutron star will be converted into strange matter during the first seconds of its existence (but note the cautionary remark in \\cite{remark}). The rest will then be transformed either by a slow burning on a time scale of a few seconds to a few minutes \\cite{olinto} or by a detonation \\cite{horvath}. For bag constants above the stability interval, we have seen that a partial transformation is still possible, but since this seems to depend heavily on the exact equation of state, one should be careful before drawing any definite conclusions.  Since a large fraction of the star is converted on a relatively short time scale, the released energy may well provide a significant contribution to the total energy of a supernova (cf.\\ Ref.\\ \\cite{horvath}).  Another investigation by Krivoruchenko and Martemyanov \\cite{kriv}, taking $\\Delta P = 0$ as a criterion for a possible transformation into strange stars by a flavor conserving phase transition, have found similar results. This is an effect not as much of equivalence of the methods used, but rather of the fact, that the high temperatures of newborn neutron stars together with the exponential in Eq.\\ (\\ref{prob}) causes the rate to be insensitive to $T$.  Another interesting feature is that if a star is born with a mass that for a given bag constant is too small for the conversion to take place even in the center, then accretion from a neighboring star, leading to a higher mass for the neutron star and thus in principle a larger transition probability, will only lead to a phase transition via thermal nucleation {\\it if} at the same time the neutron matter is heated to at least 2-5 MeV by the energy released by the accretion process or by other mechanisms, such as capture of high energy neutrinos. (It is very unlikely that a sig\\-ni\\-fi\\-cant mass can be transferred during the first second or so). Thus, one may conclude that if no mechanism for a significant heating of the star can be found, the initial mass uniquely determines the future of the star if one has to rely solely on thermal nucleation. (Other possible mechanisms that may lead to a transformation were mentioned in Sec.\\ \\ref{secintro}).  By introducing a non-zero $\\alpha_s$, a narrowing in the interesting range for $B$ was seen; both as an absolute measure and in terms of the fraction of the interval where $uds$ quark matter is stable at zero pressure.  In this work we have ignored the effect of the mass of the strange quark, which corresponds to a somewhat inadequate treatment of the equation of state, surface and curvature effects. However, even in the center of the star no more than 2-4 \\% of the quarks in the hadrons are $s$ quarks, and thus the effects during bubble nucleation are very small. As far as the surface energy is concerned, a more important effect comes from taking $\\sigma_{hadron} \\simeq (30 {\\rm MeV})^3$ (from typical nuclear mass formulae), but even here it turns out that $4\\pi r_c^2 \\sigma \\ll 8\\pi r_c \\gamma$, so that inclusion of such a term would not change our conclusions."
+    },
+    "9401/hep-ph9401262_arXiv.txt": {
+        "abstract": "We investigate the role of nuclear spin in elastic scattering of Dark Matter (DM) neutralinos from nuclei in the framework of the Minimal SUSY standard model (MSSM). The relative contribution of spin-dependent axial-vector and spin-independent scalar interactions to the  event rate in a DM detector has been analyzed for various nuclei. Within general assumptions about the nuclear and nucleon structure we find that for nuclei with atomic weights $A > 50$ the spin-independent part of the event rate $R_{si}$ is larger than the spin-dependent one $R_{sd}$ in the domain of the MSSM parameter space allowed by the known experimental data and where  the additional constraint for the total event rate $R =R_{sd} + R_{si} > 0.01$ is satisfied. The latter reflects realistic sensitivities of present and near future DM detectors. Therefore we expect equal chances for discovering the DM event either with spin-zero or with spin-non-zero isotopes if their atomic weights are $A_{1} \\sim A_{2} > 50$. ",
+        "introduction": "Analysis of the data on distribution and motion of astronomical objects within our galaxy and far beyond indicates presence of a large amount of non-luminous dark matter (DM). According to estimations, it constitutes more than $90\\%$ of the total mass of the universe  if  a mass density $\\rho$ of the universe close to the critical value $ \\rho_{crit}$ is  assumed. The exact equality $\\Omega=\\rho/\\rho_{crit}=1$, corresponding to a flat universe, is supported by naturalness arguments and by inflation scenarios. Also, in our galaxy most of the mass should be in a dark halo. Detailed models predict a spherical form for the galaxy halo and a Maxwellian distribution for DM particle velocities in the galactic frame. The mass density of DM in the Solar system should be about $\\rho \\approx 0.3 $GeV$\\cdot$cm$^{-3}$ and the DM particles should arrive at the earth's surface with mean velocities $v\\approx$ 320 km/sec, producing a substantial flux $\\Phi = \\rho\\cdot v/M$ ($\\Phi>10^{7}$cm$^{-2}$ sec$^{-1}$ for the particle mass M$\\sim$ 1 GeV). Therefore one may hope to detect DM particles directly, for instance through the elastic scattering from nuclei inside a detector.  The theory of primordial nucleosynthesis restricts the amount of baryonic matter in the universe to 10\\%. Thus a dominant component of DM is non-baryonic. The recent data by the COBE satellite \\cite{COBE} on  anisotropy in the cosmic background radiation and the theory of the formation of large scale structures of the universe lead to the conclusion that non-baryonic DM itself consists of a dominant ($70\\%$) \"cold\" DM (CDM) and smaller ($30\\%$) \"hot\" DM (HDM) component \\cite{Taylor}, \\cite{Davis}.  The neutralino ($\\chi$) is a favorable candidate for CDM. This is a Majorana ($\\chi^{c}=\\chi$) spin-half particle predicted by supersymmetric (SUSY) models.  There are four neutralinos in the minimal SUSY extension of the standard model (MSSM). They are a mixture of gauginos ($\\tilde{W}_{3}, \\tilde{B})$ and Higgsinos ($\\tilde{H}_{1,2}$), which are SUSY partners of gauge ($W_{3}, B$) and Higgs ($H_{1,2}$) bosons. The DM neutralino $\\chi$ is the lightest of them. Moreover, $\\chi$ is assumed to be the lightest SUSY particle (LSP) which is stable in SUSY models with $R$-parity conservation.  The problem of direct detection of the DM neutralino $\\chi$  via elastic scattering off nuclei has been considered by many authors and remains a field of great experimental and theoretical activity \\cite{Witt}-\\cite{Bot1}.  The final goal of theoretical calculations in this problem is the event rate $R$ for elastic $\\chi$-nucleus scattering. In general, the spin-dependent ($R_{sd}$) and spin-independent ($R_{si}$) neutralino-nucleus interactions contribute to the event rate: $R=R_{sd}+R_{si}$. $R_{sd}$ vanishes for spinless nuclei and this fact is often regarded as a reason to assert spinless nuclei to be irrelevant for the DM neutralino detection as giving a much smaller event rate. One can meet this statement in the literature. However, this is right only if the spin-dependent interaction dominates in elastic neutralino scattering off nuclei with non-zero spin.  In this paper we address the question on the role of nuclear spin in the DM neutralino detection. We investigate this problem in the framework of the MSSM. We avoid using specific nuclear and nucleon structure models but rather base our consideration on the known experimental data about nuclei and nucleon. It allows us to free the consideration of theoretical uncertainties specific for the structure models. To restrict the MSSM parameter space we use experimental constraints on SUSY-particle masses, the cosmological bound on neutralino relic abundance and the proton life-time constraint.  We have found that $R_{si}$ contribution dominates in the total event rate $R$ for nuclei with atomic weight $A>50$ in the region of the MSSM parameter space where $R =R_{sd}+R_{si}< 0.01$. The lower bound  $0.01$ is far below the sensitivity of realistic present and near future DM detectors. Therefore we can exclude the region where  $R < 0.01$ as invisible for these detectors.  We do {\\it not\\ } expect a crucial dependence of the DM event rate on the nuclear spin for detectors with target nuclei having an atomic weight larger than $50$. As a result, we expect equal chances for J~=~0 and J~$\\neq$~0 detectors to discover DM events. In particular, this conclusion supports the idea that presently operating $\\beta\\beta$-detectors with spinless nuclear target material can be successfully used for DM neutralino search. These highly developed set-ups (for a review see~\\cite{Klapdor}), operating under extremely low background conditions, use detection technology which is suitable for the DM search. ",
+        "conclusions": "In the framework of general assumptions about the nuclear and nucleon structure considering the MSSM as the basis for description of the neutralino properties we have drawn the following basic conclusions.  For sufficiently heavy nuclei with atomic weights $A > 50$ the spin-independ\\-ent event rate $R_{si}$ is larger than the spin-dependent one $R_{sd}$ if low-level signals with total event rates $R =R_{sd} + R_{si} < 0.01$ are ignored. This cut condition reflects the realistic sensitivities of the present and the near-future DM detectors.  The main practical issue is that two different DM detectors with (J~=~0, A$_1$) and with (J~$\\neq$~0, A$_2$) nuclei as target material have equal chances  to discover the DM event if  A$_1\\sim$ A$_2 >$ 50.  Another aspect of  the DM search is the investigation of the SUSY-model parameter space  from nonobservation of DM events. Apparently, in this case experiments with J~$\\neq$~0 nuclei are important since they provide new information about the SUSY model parameters from  $R_{sd}$ which is inaccessible in J~=~0 experiments.  The  results presented above were obtained in a specific SUSY-model. Therefore it is a natural question whether our basic conclusions hold for other popular SUSY-models. We plan to investigate this question in a subsequent paper."
+    },
+    "9401/astro-ph9401011_arXiv.txt": {
+        "abstract": "\\noindent We present a detailed investigation of chaotic inflation models which feature two scalar fields, such that one field (the inflaton) rolls while the other is trapped in a false vacuum state. The false vacuum becomes unstable when the magnitude of the inflaton field falls below some critical value, and a first or second order transition to the true vacuum ensues. Particular attention is paid to the case, termed `Hybrid Inflation' by Linde, where the false vacuum energy density dominates, so that the phase transition signals the end of inflation. We focus mostly on the case of a second order transition, but treat also the first order case and discuss bubble production in that context for the first time.  False vacuum dominated inflation is dramatically different from the usual true vacuum case, both in its cosmology and in its relation to particle physics. The spectral index of the adiabatic density perturbation originating during inflation can be indistinguishable from 1, or it can be up to ten percent or so higher. The energy scale at the end of inflation can be anywhere between $10^{16}$\\,GeV, which is familiar from the true vacuum case, and $10^{11}$\\,GeV. On the other hand reheating is prompt, so the reheat temperature cannot be far below $10^{11}\\,$GeV. Cosmic strings or other topological defects are almost inevitably produced at the end of inflation, and if the inflationary energy scale is near its upper limit they contribute significantly to large scale structure formation and the cosmic microwave background anisotropy.  Turning to the particle physics, false vacuum inflation occurs with the inflaton field far below the Planck scale and is therefore somewhat easier to implement in the context of supergravity than true vacuum chaotic inflation. The smallness of the inflaton mass compared with the inflationary Hubble parameter still presents a difficulty for generic supergravity theories. Remarkably however, the difficulty can be avoided in a natural way for a class of supergravity models that follow from orbifold compactification of superstrings. This opens up the prospect of a truly realistic, superstring derived theory of inflation. One possibility, which we show to be viable at least in the context of global supersymmetry, is that the Peccei-Quinn symmetry is responsible for the false vacuum. ",
+        "introduction": "\\setcounter{equation}{0} \\def\\theequation{\\thesection.\\arabic{equation}}  An attractive proposal concerning the first moments of the observable universe is that of chaotic inflation \\cite{CHAOTIC}. At some initial epoch, presumably the Planck scale, the various scalar fields existing in nature are roughly homogeneous and dominate the energy density. Their initial values are random, subject to the constraint that the energy density is at the Planck scale. Amongst them is the inflaton field $\\phi$, which is distinguished from the non-inflaton fields by the fact that the potential is relatively flat in its direction. Before the inflaton field $\\phi$ has had time to change much, the non-inflaton fields quickly settle down to their minimum at fixed $\\phi$, after which inflation occurs as $\\phi$ rolls slowly down the potential.  Two possibilities exist concerning the minimum into which the non-inflaton fields fall. The simplest possibility is that it corresponds to the true vacuum; that is, the non-inflaton fields have the same values as in the present universe. Inflation then ends when the inflaton field starts to execute decaying oscillations around its own vacuum value, and the hot Big Bang (`reheating') ensues when the vacuum value has been achieved and the decay products have thermalised. This is the usually considered case, which has been widely explored. The other possibility is that the minimum corresponds to a false vacuum, with non-zero energy density. This case may be called {\\it false vacuum inflation}, and is the subject of the present paper.  There are two fundamentally different kinds of false vacuum inflation, according to whether the energy density is dominated by the false vacuum energy density or by the potential energy of the inflaton field. (For simplicity we discount for the moment the intermediate possibility that the two contributions are comparable, though it will be dealt with in the body of the paper.) In all cases the false vacuum exists only when the value of the inflaton field is above some critical value. If the false vacuum energy dominates, a phase transition occurs promptly when the inflaton field falls below the critical value, causing the end of inflation and prompt reheating. The result is a new model of inflation which is dramatically different from the usual one, and at least as attractive. It was first studied by Linde who termed it `Hybrid Inflation', and it is the main focus of the present paper. The phase transition may be of either first or second order. A first-order model of false vacuum dominated inflation has been considered by Linde \\cite{Linde90} and (with minor differences but more thoroughly) by Adams and Freese \\cite{AdamsFreese}. A second-order model has been discussed by Linde \\cite{LIN2SC,LIN2SC2} and explored in a preliminary way by Liddle and Lyth \\cite{LL2} and by Mollerach, Matarrese and Lucchin \\cite{MML}. As far as we know these are the only references in the literature to false vacuum dominated inflation with Einstein gravity. Related models have been considered at some length in the context of extended gravity theories \\cite{LaSteinhardt,extinf,Amendola}; although such theories can be recast as Einstein gravity theories by a conformal transformation, the resulting potentials are of a different type and this case is excluded from the present paper.  The opposite case where the false vacuum energy is negligible (inflaton domination) is indistiguishable from the true vacuum case for couplings of order unity, though a variety of exotic effects can occur for small couplings. This case has been studied by several authors \\cite{Kofman,vishniac,kofpog,yokoyama,kbhp,sbb,lyth90,hodpri,nagasawa}, and in the present paper it is treated fairly briefly.  {}From the viewpoint of cosmology, false vacuum dominated inflation differs from the usual true vacuum case in three important respects. \\begin{enumerate} \\item The spectral index $n$ of the adiabatic density perturbation is typically very close to the scale invariant value 1, and is in any case greater than 1. This is in contrast with other working models of inflation, where one typically finds $n<1$, viable models covering a range from perhaps $n \\simeq 0.7$ up to $n \\simeq 1$ \\cite{LL2}. We shall however note that the extent to which $n$ can exceed unity is quite limited, contrary to claims in Refs.~\\cite{LIN2SC2,MML}. \\item Topological defects generally form at the {\\it end} of inflation, in accordance with the homotopy groups of the breaking of the false vacuum to degenerate states, provided that these groups exist. The defects may be of any type (domain walls, gauge or global strings, gauge or global monopoles, textures or nontopological textures). \\item Reheating occurs promptly at the end of inflation. In the simple models that we have explored, this means that the reheat temperature is at least $10^{11}$\\,GeV. One consequence is that a long lived gravitino must be either rather heavy ($m\\gsim 1$\\,TeV) or extremely light, so as not to be overproduced \\cite{kawashi}. \\end{enumerate}  False vacuum dominated inflation is also very different from the true vacuum case from the viewpoint of particle physics. Sticking to the chaotic inflation scenario already described, let us consider as a specific example the inflationary potential \\begin{equation} V(\\phi)=V_0+ \\frac12 m^2\\phi^2+\\frac14\\lambda \\phi^4 \\,. \\end{equation} where $V_0$ is the false vacuum energy density. Consider first the true vacuum case, where $V_0$ vanishes. Inflation occurs while $\\phi$ rolls slowly towards zero, and it ends when $\\phi$ begins to oscillate, which occurs when $\\phi$ is of order the Planck mass. In order to have sufficiently small cosmic microwave background (cmb) anisotropy, one needs $m\\lsim 10^{13}\\,$GeV and $\\lambda\\lsim 10^{-12}$, with one or conceivably both of these limits saturated if inflation is to actually generate the observed anisotropy (and a primeval density perturbation leading to structure formation). To achieve the small $\\lambda$ in a natural way one should invoke supersymmetry. As long as one sticks to global supersymmetry this presents no problem, but there are sound particle physics reasons for invoking instead local supersymmetry, which is termed supergravity because it automatically includes gravity. In the context of supergravity, the fact that $\\phi$ is of order the Planck mass during inflation is problematical, because in this regime it is difficult to arrange for a sufficiently flat potential.  As will become clear, things are very different in the false vacuum case. One still needs to have $\\lambda$ very small, and will still therefore wish to implement inflation in the context of supergravity. But now $\\phi$ is far below the Planck scale during inflation (after the observable universe leaves the horizon which is the cosmologically interesting era). As a result it becomes easier to construct  a viable model of inflation, though the smallness of $m$ in relation to the inflationary Hubble scale $H$ still presents a severe problem for generic supergravity theories. Remarkably though, it turns out that among the class of supergravity models emerging from orbifold compactifications of superstring theory, one can find a large subset for which this problem disappears. As a toy model, we will see how things work out with a specific choice for the perturbative part of the superpotential.  Another crucial difference concerns the mass $m$. In contrast with the true vacuum case, the cmb anisotropy does not determine $m$ in the vacuum dominated case, but rather determines $V_0$ as a function of $m$. The value $m\\sim 10^{13}\\,$GeV that obtains in the true vacuum case is allowed as an upper limit, but $m$ can be almost arbitrarily small and it is natural to contemplate values down to at least the scale $m\\sim 100\\,$GeV. The value of $m$ chosen by nature might be accessible to observation because it determines the spectral index $n$; if $m$ is within an order of magnitude or so of its upper limit $n$ is appreciably higher than 1, whereas if it is much lower $n$ is indistinguishable from 1. In the superstring motivated models mentioned earlier, the first case probably obtains if the slope of the inflationary potential is dominated by one-loop corrections coming from the Green-Schwarz mechanism, in which case the value of $n$ is determined by the orbifold. This would open up the interesting possibility that observations of the cmb anisotropy and large scale structure provide a window on superstring physics.  The opposite case $m\\sim100\\,$GeV is also interesting. Supersymmetric theories of particle physics typically contain several scalar fields with this mass. The corresponding false vacuum energy scale $V_0^{1/4}\\sim 10^{11}\\,$GeV also appears in particle physics, as that associated with Peccei-Quinn symmetry, a global $U(1)$ symmetry which is perhaps the most promising explanation for the observed CP invariance of the strong interaction. This same symmetry provides the axion, which is one of the leading dark matter candidates, and the possibility that it might in addition provide the false vacuum for inflation is to say the least interesting. We explore this possibility in the context of global supersymmetry and find that it can easily be realised there. We have not gone on to explore it in the context of supergravity, but there seems to be no reason why it should not be realised within the context of the superstring derived models considered earlier.  As will be clear from this introduction, the present work is expected to be of interest to a very wide audience, ranging from observational astronomers to superstring theorists. With this in mind we have tried to keep separate the part of the paper that discusses the phenomenology of the false vacuum inflation models, and the part that relates these models to particle physics.  The outline of the paper is as follows. Section \\ref{INFL} introduces the specific second-order model upon which most of our discussion shall be focussed. We analyse the inflationary dynamics and density perturbation constraints by a combination of analytic and numerical methods to delineate the observationally viable models. Section \\ref{TOP} then takes our attention onto the formation of topological defects, which (almost) inevitably form at the end of inflation. Their possible existence constrains the models, and there is the further opportunity of a reconciliation of structure-forming defects with inflation. In Section \\ref{SUGR} we try to realise the model in the contexts of global supersymmetry, supergravity and superstring derived supergravity. In Section \\ref{MODELS} we consider the related first-order model which also indicates the link with extended inflation models. Section \\ref{CONC} summarises the paper. ",
+        "conclusions": "\\label{CONC} \\setcounter{equation}{0} \\def\\theequation{\\thesection.\\arabic{equation}}  In conclusion, models of inflation based on Einstein gravity, but driven by a false vacuum, offer a range of new possibilities for both theory and phenomenology.  On the particle physics side, we have shown how false vacuum inflation points to new possibiities for model building. In particular, we have shown that it can occur in a class of supergravity models implied by orbifold compactification of superstrings. One outcome of that discussion was the intriguing possibility of obtaining a handle on the superstring orbifold, through the fact that one-loop corrections might be the dominant effect determining the spectral index. Much remains to be done of course. For instance, although we have exhibited a toy model for the scalar field sector of the string derived supergravity theory, we have made no attempt to put it in the context of a realistic model involving other fields as well. In particular we have not tried to extend to supergravity the identification of the false vacuum with that of Peccei-Quinn symmetry, which we found was both viable and attractive in the context of global supersymmetry.  In terms of direct cosmological phenomenology, false vacuum dominated inflation offers the unusual option of a spectral index for the density perturbations exceeding unity, though we have demonstrated that with the COBE normalisation the deviation can only be rather modest with a plausible maximum of around $n = 1.14$. There is however additional interest in that one expects topological defects to form as the false vacuum decays; because essentially all the energy density is available to go into the defect fields, the energy available is much greater than in usual models where reheating is required first, redistributing the energy into a large number of fields. Because of this, structure-forming defects are comfortably compatible with our inflation model when the masses are towards the top of their allowed ranges.  We have also made a preliminary investigation of the details of the phase transition in different regimes, though much remains to be done. For a second-order phase transition, results already exist in the literature describing the inflaton dominated regime. We have demonstrated that, barring very weak couplings, the phase transition proceeds very rapidly in the vacuum dominated regime, but have been unable to develop a solid understanding of the statistics of the defects produced in such a transition. In the first-order case, where the transition completes via bubble nucleation, we have gone on to calculate the bubble distribution and the constraints upon it. We note that first-order inflation models based on Einstein gravity are generally easier to implement than those of the extended inflation type.  That one can have both structure-forming topological defects and inflation raises a host of possible structure formation scenarios, as one could choose to utilise only one of these two or a combination of the two. It is believed \\cite{Albert} that for a given size of density perturbation (i.~e.~perturbation in the gravitational potential), defects give a larger microwave background temperature anisotropy, by a factor of a few. One could therefore arrange for defects to be the source of a component of the COBE signal while having only a modest effect on structure formation; alternatively one could aim to have inflation and defects contributing roughly equally to structure formation in which case the defects would be predominant in the microwave background. It is conceptually (and calculationally) preferable to take the option of using only one source, lowering the energy scale of the other to make its effects negligible, but one should be aware that the required scales of the two are similar, and should a realistic model along our suggested lines be devised it would not be a particular surprise should both contributions have a role to play."
+    }
+}