1: \documentclass[]{aastex}
2: \usepackage{emulateapj5,latexsym,graphics,epsfig}
3: \def \be{\begin{equation}}
4: \def \ee{\end{equation}}
5: \def \bea{\begin{eqnarray}}
6: \def \eea{\end{eqnarray}}
7: \def\etal{{et al.\ }}
8: \slugcomment{Submitted to ApJ ?}
9:
10: \begin{document}
11:
12: \title{Dust sputtering by Reverse Shocks in Supernova Remnants}
13:
14: \author{Biman B. Nath$^1$, Tanmoy Laskar$^{1,2}$ and J. Michael Shull$^3$}
15: \affil{$^1$Raman Research Institute, Sadashivanagar, Bangalore 560080, India\\
16: $^2$St. Stephen's College, Delhi 110007, India\\
17: $^3$CASA, Department of Astrophysical and Planetary Sciences, University
18: of Colorado, Boulder, CO 80309-0389, USA}
19: \email{biman@rri.res.in, mshull@casa.colorado.edu}
20:
21: \shorttitle{Dust sputtering by reverse shock in SN}
22: \shortauthors{B. Nath, T. Laskar, J. M. Shull}
23:
24: \begin{abstract}
25: We consider sputtering of dust grains, believed to be formed in cooling
26: supernovae ejecta, under the influence of reverse shocks. In the regime
27: of self-similar evolution of reverse shocks, we can follow the evolution
28: of ejecta density and temperature analytically as a function
29: of time in different parts of the ejecta, and calculate the sputtering
30: rate of graphite and silicate grains embedded in the ejecta as they
31: encounter the reverse shock. Through analytic (1D) calculations,
32: we find that a fraction of dust
33: mass ($ 1\hbox{--}20$\% for silicates and %$\le 5$\% for
34: graphites)
35: can be sputtered by reverse shocks, the fraction varying with the
36: grain size distribution and the
37: steepness of the density profile of the ejecta mass. It is expected that
38: many more grains will get sputtered in the region between the forward and
39: reverse shocks, so that our analytical results provide a lower limit to
40: the destroyed fraction of dust mass.
41: \end{abstract}
42:
43: \keywords
44: {ISM : Supernova Remnants, ISM : Dust, Extinction, Galaxies : High-Redshift}
45:
46: \section{Introduction}
47: Understanding the chemical history of heavy elements (``metals")
48: in the interstellar medium (ISM) of galaxies is an important, but
49: difficult undertaking. Because refractory elements are found to
50: be depleted from the gas phase (locked into grains), they provide
51: a substantial reservoir of coolants in the solid-state phase.
52: Interstellar grain surfaces are believed to furnish formation sites
53: for molecular hydrogen (H$_2$), which is both a coolant and the
54: starting point for a rich cloud chemistry. Grains also provide a
55: critical transfer mechanism for reprocessing UV/O starlight into
56: far-infrared and sub-mm emission from interstellar molecular clouds
57: and high-redshift galaxies. In the high-redshift universe, the
58: first dust grains may influence the thermodynamics of the
59: primordial gas and thereby control rates of star formation.
60:
61: Recent observations of dust grains at high
62: redshift have been puzzling in the context of their formation.
63: Several observations of damped Ly$\alpha$ systems (Pettini \etal 1994; Ledoux,
64: Bergeron \& Petitjean 2002) have shown evidence of dust grains in them. Also,
65: thermal emission (at $1.2$ mm) from dust in high redshift ($z >6$) QSOs in
66: the Sloan Digital Sky Survey (SDSS)
67: have been reported (Bertoldi \etal 2003).
68: The implied dust mass of $\sim 10^8$ M$_\odot$
69: within a Gyr after the Big Bang
70: appears difficult to explain with traditional
71: models of grain formation in evolved low mass stars (Dunne \etal 2003), from
72: which grains are thought to be
73: transported to the interstellar medium (ISM)
74: through stellar winds (Whittet 1992).
75:
76: Recent models have, therefore, focused on other sites
77: of dust grain formation, such as ejecta of core-collapse
78: supernova explosions that
79: can occur on shorter time scales than that of the evolution of low mass stars
80: (Dwek \& Scalo 1980). Core-collapse supernovae are thought to be capable of
81: synthesizing a significant amount of % $0.2\hbox{--}4$ M$_{\odot}$ of
82: dust in the SN ejecta.
83: Applying a theory of nucleation and grain growth developed by
84: Kozasa \& Hasegawa (1987), Kozasa \etal (1989; 1991)
85: and Todini \& Ferrara (2001)
86: have studied grain formation in expanding SN ejecta, the latter estimating
87: the dust mass formed $\sim 0.1\hbox{--}0.3$ M$_{\odot}$. Nozawa \etal (2003)
88: have extended it to the
89: case of zero-metallicity SNe for population III stars (see also Schneider,
90: Ferrara \& Salvaterra 2004).
91:
92: Although the SN ``dust factory model'' would explain the large
93: observed dust masses in high-redshift galaxies, observations of
94: dust in nearby SNe appear inconsistent with this model, falling
95: factors of 10--100 short of the required amounts ($\sim0.2M_{\odot}$;
96: see above).
97: As an example,
98: in a review of Type~II SN~2003gd (NGC~628),
99: Sugerman et al.\ (2006) state that ``radiative transfer models show
100: that up to 0.02 solar masses of dust has formed within the ejecta".
101: However, a recent study (Meikle et al.\ 2007) of the same remnant
102: concludes that the mid-IR flux ``is consistent with emission from
103: $4 \times 10^{-5}~M_{\odot}$ of newly condensed dust in the ejecta".
104: In a young, oxygen-rich SNR in the SMC, Stanimirovic et al.\ (2005)
105: found only $10^{-3}~M_{\odot}$ of hot dust ($T_d \approx 120$~K).
106: %In the Large Magellanic Cloud, four core-collapse supernovae were
107: %studied at 24 and 70~$\mu$m by {\it Spitzer} (Williams et al.\ 2006),
108: %but with inferred dust masses only 0.01--0.1~$M_{\odot}$.
109: The claimed sub-mm detection (Dunne et al.\ 2003) of $\sim3~M_{\odot}$
110: of cold dust toward the Cas~A remnant was later shown to arise from
111: interstellar dust in an adjacent molecular cloud (Krause et al.\ 2004).
112:
113:
114: Observational tests of predictions made by models of
115: SNe are difficult, involving far-IR and
116: sub-mm observations of young SNRs. One must disentangle far-IR
117: backgrounds, separate newly synthesized dust from circumstellar
118: and interstellar dust, and understand the role of SNR reverse
119: shocks, which can destroy newly synthesized dust in high-speed ejecta.
120: A case in point is SN 1987A.
121: Dust formation was inferred in ejecta of SN~1987A, but only
122: $10^{-3}~M_{\odot}$ was detected (Dwek 2006; Dwek \& Arendt 2007).
123: Additional dust is almost certainly present, because of observed changes
124: in the optical and bolometric fluxes and emission-line asymmetries in
125: the red and blue wings (McCray 1993, 2007). Moreover, the reverse shock
126: has not yet reached the interior of the supernova debris in SN~1987A.
127: These inner ejecta are most prone to dust formation, since they are
128: cold, dense, and metal-enriched.
129:
130: A challenging theoretical issue is whether high-velocity dust
131: ($V_{\rm ej} \geq 1000$ km~s$^{-1}$) can survive the shocks,
132: thermal sputtering, and grain-grain collisions, as the ejecta
133: are slowed down by the surrounding ISM.
134: A goal of this paper is to determine how much dust survives
135: the SN event and is incorporated into the surrounding ISM. The relative
136: speed of reverse shocks with respect to the expanding ejecta can be very large,
137: and can raise the temperature of the ejecta to the extent of destroying the
138: nascent dust grains in it. It is therefore important to study the effect
139: of reverse shocks in detail.
140:
141: Recently, Nozawa \etal (2007) have considered this problem and came to
142: the conclusion, with the help of hydrodynamical simulation, that reverse
143: shocks can destroy a fraction of dust mass of order $20\hbox{--}100$\%
144: depending on the ambient gas density and explosion energy. Bianchi \&
145: Schneider (2007) also used numerical simulations and semi-analytical methods to
146: conclude that only a small amount of dust mass survives; e.g.,
147: $\sim 7$\% dust mass survives in the case of a SN with
148: a progenitor of mass $20 \,$ M$_{\odot}$ and ambient particle density of
149: $\sim 1$ cm$^{-3}$. Nozawa \etal (2007) have included the motion
150: of dust relative to gas within the SN remnant and calculated the evolution
151: of dust grains over a period spanning the radiative phase ($\sim 10^5 \hbox{--}
152: 10^6$ yr), whereas Bianchi \& Schneider (2007) neglected the effect
153: of motion of grains on dust destruction and calculated dust evolution
154: up to the non-radiative phase ($\le 10^5$ yr).
155:
156:
157: We study in this paper a simple and analytically tractable problem, namely
158: the effect of reverse shocks for a power-law
159: mass distribution of the ejecta, in the
160: regime of self-similar evolution of both forward and reverse shocks. Although
161: the simple formalism does not allow us to study several important
162: processes--- including motion of dust grains and its effect on evolution
163: of dust grains--- our approach allows one to segregate the processes
164: of destruction within the reverse shock itself from those occuring between
165: forward and reverse shocks (which can be dealt with in numerical
166: simulations involving grain motion).
167:
168:
169: \section{Dust in supernova remnants}
170: We first summarize the results of the theoretical predictions of dust grain
171: formation in SNRs. Todini \& Ferrara (2001) calculated the formation of
172: different types of dust grains--- amorphous carbon (AC), silicates (enstatites
173: : $MgSiO_3$, forsterites : $Mg_2SiO_4$) and iron-bearing magnetites ($Fe_3O_4$)
174: and others, in ejecta with different metallicities. They found that AC
175: grains are generally of larger size, in the range of $\sim 0.01\hbox{--}0.1 \,
176: \mu$~m, whereas silicates are produced with sizes $\sim 10^{-3} \, \mu$~m.
177: Nozawa \etal (2003) reached similar conclusions for Pop III SN events, with
178: sizes of newly synthesized grains spanning a range of three orders of
179: magnitude, with maximum grain size being less than $\sim 1 \, \mu$m.
180:
181: The dust grain parameters in the ISM are inferred to be somewhat
182: different from these
183: estimates. The extinction observations are satisfactorily reproduced by
184: assuming mostly of two components--- graphite and silicate--- with
185: a common power-law size distribution, $dn/da \propto a^{-3.5}$, truncated
186: at a minimum size $a_l\sim 5 \times 10^{-3} \, \mu$m and a maximum size
187: $a_m \sim 0.25 \, \mu$m (Draine 2003, and references therein). There
188: also appears to be a population of very small grains with PAH composition.
189: Nozawa \etal (2003) found for grains synthesized in Pop III SN ejecta
190: a size distribution with index $-2.5$ for smaller grains and with $-3.5$ for
191: larger grains (with the grain size at crossover point depending on supernovae
192: models). We use these two power laws to bracket the possible size
193: distributions of grains synthesized in SN ejecta, and assume that grains
194: are formed in the range $10^{-7}\hbox{--}3 \times 10^{-5}$ cm (see Figure 11 of
195: Nozawa \etal (2003). We also consider the case of an upper limit of
196: $10^{-6}$ cm and compare our results for both upper limits.
197:
198: It is instructive to estimate the destruction time scales before we calculate
199: the effect of reverse shock in detail.
200: The sputtering time scale for dust (graphite,
201: silicate and iron) grains of size $a$
202: is approximately $\sim 10^6 \, ({a \over 1 \, \mu {\rm
203: m}} ) \, ({n_e
204: \over 1 \,
205: {\rm cm}^{-3}})^{-1}$ yr, for ambient temperatures larger than $10^6$ K
206: (Draine \& Salpeter 1979), where $n_e$
207: is electron density. For grains with size $\sim 0.1\hbox{--}0.3$ $\mu m$
208: (which incidentally contain most of the dust mass for a $a^{-3.5}$
209: distribution), and for $n_e\sim 1\hbox{--}10$ $cm
210: ^{-3}$, the sputtering time scale is $\sim 2 \times 10^{4\hbox{--}5}$ yr.
211: %The grain temperature (for gas density $n=1 \hbox{--}10$ cm$^{-3}$ and gas
212: %temperature $T\sim 10^7$ K) is of order $\sim 130$ K (Dwek 1987).
213:
214:
215:
216: \section{Reverse shocks}
217: We now briefly discuss the self-similar evolution of
218: forward and reverse shocks in SN (see Truelove \& McKee 1999, hereafter
219: referred to as TM99, for further
220: details).
221:
222: \subsection{The ejecta}
223: The parameters used to describe a supernova remnant which affect its
224: evolution are
225: the energy of the explosion ($E_{ej}$),
226: the mass contained in the expanding ejecta ($M_{ej}$),
227: the maximum velocity of the material within the ejecta ($v_{ej}$), and
228: the density of the ISM ($\rho_{0}$).
229: Denoting the time of the explosion as $t=0$, we construct our
230: initial conditions at a later time, still early in the history of the remnant,
231: assuming the ejecta to have already expanded to a radius $R_{ej}$
232: without any deceleration due to the ambient medium.
233: If we consider SN to be a point explosion, then the radial velocity profile
234: will be linear as ejecta with velocity $v$ will travel out to a distance $vt$ :
235: \begin{equation}
236: v(r) = \left\{ \begin{array} {ll}
237: \frac{r}{t}, & \mbox{$r < R_{ej}$} \\
238: 0, & \mbox{$r > R_{ej}$}
239: \end{array}
240: \right .
241: \end{equation}
242: This profile becomes flatter as time progresses and the faster-moving
243: ejecta are flung out to greater distances.
244: The corresponding radial density distribution of the ejecta matter is given by:
245: \begin{equation}
246: \rho(r) = \left\{ \begin{array} {ll}
247: \rho_{ej}(v,t) \equiv \frac{M_{ej}}{v_{ej}^3}
248: f(\frac{v}{v_{ej}})t^{-3}, & \mbox{$r < R_{ej}$} \\
249: \rho_{0}, & \mbox{$r > R_{ej}$}
250: \end{array}
251: \right.
252: \label{eq:den}
253: \end{equation}
254: where $f(v/v_{ej})$ is the time-independent structure function of the density
255: profile and the $t^{-3}$ term arises from the free expansion of the ejecta.
256: We define a dimensionless parameter to label shells of the ejecta starting
257: from the centre outwards as
258: \be
259: w = \frac{v}{v_{ej}}, 0 \leq w \leq 1
260: \ee
261: Following TM99 we consider the situation when
262: \textit{f} is given by a power law in \textit{w}:
263: \be
264: f(w) = \left\{ \begin{array} {ll}
265: f_{0}, & \mbox{$0 \leq w \leq w_{c}$} \\
266: f_{n}w^{-n}, & \mbox{$w_{c} \leq w \leq 1$}
267: \end{array}
268: \right.
269: \ee
270: where we have separated the ejecta into a uniform density core region
271: and a power-law envelope region. The continuity of \textit{f} at $w_{c}$
272: and the normalization of $\rho$ translate to expressions for parameters $f_{0}$
273: and $f_{n}$ in terms of the free parameters $w_{c}$ and \textit{n}:
274: \bea
275: f_{0} &=& f_{n}w_{c}^{-n} \, \nonumber\\
276: f_{n} &=& \frac{3}{4\pi} \left[ \frac{1-n/3}{1-(n/3)w_{c}^{3-n}}\right] \,.
277: \eea
278:
279: \subsection{Forward and reverse shocks}
280: As the freely expanding ejecta come into contact with the ambient medium, with
281: the contact discontinuity moving at $v_{ej}$, a speed much larger than the
282: sound speed in the ambient medium, a forward shock is set up in the
283: surrounding gas
284: which propagates into the medium. The ejecta at the contact surface decelerates
285: suddenly as it hits the ambient gas, and consequently, a reverse shock is
286: set up that propagates inwards through the ejecta. While the mass of this
287: matter swept by the forward shock is less than the total mass of the ejecta
288: driving the shock, the remnant is said
289: to be in the \textit{Ejecta Dominated} or ED stage. It is in this stage, before
290: the reverse shock has attained a significant velocity, that the ejecta can be
291: approximated as a spherical piston freely expanding into the ambient gas. When
292: the mass swept by the forward shock becomes approximately equal to the ejecta
293: mass, the corresponding late-time limit is known as the
294: \textit{Sedov-Taylor} or ST stage of the evolution of the remnant.
295:
296: We follow TM99 to follow the evolution of the reverse shock analytically.
297: Let $R_{f}$ and $R_{r}$ denote the radii of the forward and reverse
298: shocks, and $v_{f} \equiv dR_{f}/dt$ and
299: $v_{r} \equiv dR_{r}/dt$ be the velocities of the forward and reverse shock
300: waves respectively, all in the rest frame of the unshocked ambient medium.
301: We have the velocity of the unshocked ejecta as seen by the reverse shock as
302: it propagates into the ejecta:
303: \be
304: \tilde{v}_{r} \equiv v(R_{r},t) - v_{r}(t) = \frac{R_{r}(t)}{t} - v_{r}(t)\,.
305: \ee
306: Let \textit{$\phi$} be the ratio of the pressures behind the reverse and
307: forward shock waves and \textit{l} be the ratio of the radii of the two shocks
308: at a given time:
309: \be
310: \phi(t) \equiv \frac{\rho_{ej}(v_{r},t)\tilde{v}_{r}^2(t)}{\rho_{0}v_{f}^2(t)}
311: \,,\quad
312: l(t) \equiv \frac{R_{f}(t)}{R_{r}(t)} \,, \quad
313: w_f(t) \equiv \frac{R_f(t)/t}{v_{ej}} \,.
314: \ee
315:
316:
317: It is useful to express various parameters in dimensionless forms in the
318: units of characteristic values of these variables. so that,
319: $R_\ast=R/R_{ch}$,
320: $t_\ast=t/t_{ch}$, and $v_\ast=v/v_{ch}$. For example, the
321: characteristic scales of length, time and velocity are given by,
322: \bea
323: R_{ch}&=& M_{ej}^{1/3} \rho_0^{-1/3} \approx 5.25 \, n_0^{-1/3} \,
324: \, {\rm pc} \,, \nonumber\\
325: t_{ch}&=& E^{-1/2} M_{ej}^{5/6} \rho_0^{-1/3}
326: \approx 1.66 \times 10^3 \, n_0^{-1/3} \,\, {\rm yr} \,, \nonumber\\
327: v_{ch}&=& R_{ch}/t_{ch} \approx 3.16 \times 10^3 \, {\rm km} \,\,
328: {\rm s}^{-1}\,.
329: \label{eq:char}
330: \eea
331: (We have assumed $\rho _0=1.4 n_0 m_p$ for He/H$=0.1$ by number.)
332: We have estimated the numerical values for the case $E=10^{51}$ erg, $M_{ej}=
333: 10^{34}$ g and $n_0=1$ cm$^{-3}$.
334: We will therefore write
335: $R^\ast=R/R_{ch}$,
336: $t^\ast=t/t_{ch}$, and $v^\ast=v/v_{ch}$, and the
337: radius of the reverse shock, in particular, as
338: $R_r^\ast\equiv R_r/R_{ch}$.
339:
340: \subsubsection{ED (Expansion dominated) stage}
341: The values of $\phi$ and $l$ approach constant values as $t \rightarrow 0$ if
342: the solutions remain self-similar in the same limit. We write $\phi(t) \simeq
343: \phi(0) \equiv \phi$ and $l(t) \simeq l(0) \equiv l$, and
344: also $w_{f}(0) = l$. The solution to reverse shock evolution exists in the form
345: of two branches, which we label the \textit{envelope} ($w_{c} \leq
346: w_{f}/l_{ED} \leq 1$) and \textit{core} ($0 \leq w_{f}/l_{ED} \leq w_{c}$)
347: branches. Denoting $\alpha \equiv \frac{E}{(1/2)M_{ej}v_{ej}^{2}}$,
348: the evolution of the reverse shock is given
349: by,
350: \bea
351: &&t^\ast(R_{r}^\ast) = \left(\frac{\alpha}{2}\right)^{1/2}R_{r}^\ast \times
352: \nonumber\\
353: &&
354: \left\{ \begin{array} {ll}
355: \left[1 + \left(\frac{n-3}{3}\right)\left(\frac{\phi}{f_{n}}\right)^{1/2} l
356: R_{r}^{\ast 3/2}\right]^{-2/(3-n)}
357: , \\%& \mbox{C} \\
358: \left[w_{c} - \left(\frac{\phi}{lf_{0}}\right)^{1/2} \left\{l^{3/2}R
359: _{r}^{\ast 3/2} - \frac{f_{n}^{1/2}}{3-n}\left(1 - w_{c}^{(3-n)/2}\right)
360: \right\}\right]^{-2/3}
361: , %& \mbox{E}
362: \end{array}
363: \right.
364: \label{eq:ednl3_1}
365: \eea
366: where the first part corresponds to the case $w_{c} \leq w_{f}/l \leq 1$
367: and the second part, to $0 \leq w_{f}/l \leq w_{c}$. The corresponding
368: velocity of reverse shock in these two cases are,
369: \bea
370: && v^\ast(R_{r}^\ast) = \left(\frac{2}{\alpha}\right)^{1/2} \times \nonumber\\
371: && \left\{ \begin{array} {ll}
372: \frac{
373: \left[1 - \left(\frac{3-n}{3}\right)\left(\frac{\phi}{f_{n}}\right)^{1/2}
374: lR_{r}^{\ast 3/2}\right]^{(5-n)/(3-n)}}{\left[1+\frac{n}{3}\left(\frac{\phi}
375: {f_{n}}\right)^{1/2}lR_{r}^{\ast 3/2}\right]}
376: , & \\%\mbox{$w_{c} \leq w_{f}/l \leq 1$} \\
377:
378: \frac{\left[w_{c} - \left(\frac{\phi}{lf_{0}}\right)^{1/2} \left\{l^{3/2}
379: R_{r}^{\ast 3/2} - \frac{f_{n}^{1/2}}{3-n}\left(1 - w_{c}^{(3-n)/2}\right)
380: \right\}\right]^{5/3}}{\left[w_{c} + \frac{f_{n}^{1/2}}{3-n}\left(1 -
381: w_{c}^{(3-n)/2}\right)\right]}
382: , & %\mbox{$0 \leq w_{f}/l \leq w_{c}$}
383: \end{array}
384: \right.
385: \eea
386: The relative velocity of the reverse shock in the frame of reference of
387: the expanding ejecta, for the case when the reverse shock is in the
388: envelope region, is given by,
389: \be
390: \tilde{v}_r^\ast(R_{r}^\ast) = %
391: \sqrt{{2 \phi \over \alpha f_{n}}}%\left(\frac{2\phi}{\alpha f_{n}}\right)^{1/2}
392: lR_{r}^{\ast 3/2} \,
393: \left(
394: \frac{
395: \left[1 - \left(\frac{3-n}{3}\right)\left(\frac{\phi}{f_{n}}\right)^{1/2}
396: lR_{r}^{\ast 3/2}\right]^{2/(3-n)}}{\left[1+\frac{n}{3}\left(\frac{\phi}{f_{n}}
397: \right)^{1/2}lR_{r}^{\ast 3/2}\right]}\right) \,.
398: \label{eq:ednl3_2}
399: \ee
400: Note that, since $w_f/l =\frac{R_r(t)/t}{v_{ej}}$
401: describes the reverse shock, we will define it as `$w$' and use it to label
402: shells in the ejecta.
403:
404: This solution is valid (in the ED stage) only as long as
405: the reverse shock remains in the envelope $w_{core}\leq w \leq 1$.
406: In the case of $n < 3$ ejecta, there is no need for a core and we can take the
407: limit $w_{core} \rightarrow 0$. For $n > 3$ however, we must have a ``flat
408: profile'' core to keep the integrated mass finite.
409: For the steep index scenario, therefore, the functional form of the solution
410: will change when the reverse shock reaches the core and we will return to
411: this point later.
412:
413: \subsubsection{ST (Sedov-Taylor) stage}
414: We now describe the Sedov-Taylor stage, which is the late-time evolution limit
415: of the non-radiative phase of a supernova remnant.
416: Following TM99, we assume that the shocks trace a constant acceleration path
417: after transition to the ST stage at $t_{ST}$, which (approximately) begins
418: when the mass of the ambient medium shocked by the forward shock equals the
419: mass ejected by the supernova. Thus the velocity of the oncoming ejecta in the
420: frame of the reverse shock is
421: \be
422: \tilde{v}_{r}^{*}(t^{*}) = \tilde{v}_{r,ST}^{*} + \tilde{a}_{r,ST}^{*}(t^{*}
423: - t_{ST}^{*}) \,,
424: \label{eq:stnl3_2}
425: \ee
426: where $\tilde{a}_{r,ST}$ is the constant acceleration and $\tilde{v}_{r,ST}$
427: is the velocity of the ejecta w.r.t the reverse shock at $t_{ST}$.
428: Note that $\tilde{v}_{r} = -td(R_{r}/t)/dt$. Using this and integrating the resulting expression from $t_{ST}$ to $t$ gives
429: \be
430: R_r ^\ast (t^\ast) = t^\ast \left[\frac{R_{r,ST}^\ast} {t_{ST}^\ast} -
431: \tilde{a}_{r,ST}^\ast (t^\ast -t_{ST}^\ast ) - (\tilde{v}_{r,ST}^\ast -
432: \tilde{a}_{r,ST}^\ast t_{ST}^\ast )
433: ln\left(\frac{t^\ast }{t_{ST}^\ast }\right)\right] \,.
434: \label{eq:stnl3_1}
435: \ee
436:
437: For the $n < 3$ case, we therefore have an implicit solution for the reverse
438: shock position in the ED stage $t < t_{ST}$ and an explicit solution in the
439: form of a uniform acceleration description in the ST stage.
440: For the $n > 5$ case, we again have an implicit solution in the ED stage at
441: least while $t < t_{core}$. In fact, an explicit solution for $R_{r}^\ast (t^
442: \ast )$ can be determined from the implicit solution in the ED stage for
443: $n > 5$ remnants. We have (see equation (74) of TM99),
444: \be
445: t^\ast (R_{r}^\ast ) = \left(\frac{\alpha}{2}\right)^{1/2}R_{r}^\ast \,
446: \left[1 + \left(\frac{n-3}{3}\right)\left(\frac{\phi}{f_{n}}\right)^{1/2}
447: lR_{r}^{\ast 3/2}\right]^{-2/(3-n)} \,.
448: \ee
449: In this expression, we can take the limit $w_{core} \rightarrow 0$ by
450: allowing $v_{ej} \rightarrow \infty$, the ejecta energy $E$ remaining finite.
451: In this limit $f_{n} \rightarrow 0$ and when the second term in the brackets
452: becomes much larger than unity, we have
453: \be
454: R_{r}^\ast (t^\ast ) = \left[\frac{27}{4\pi}\frac{1}{n(n-3)l^{2}\phi}
455: \left\{\frac{10}{3}\left(\frac{n-5}{n-3}\right)\right\}^{\frac{n-3}{2}}\right]
456: ^{1/n} \bigl ( t^{\ast} \bigr ) ^{ \frac{n-3}{n}} \,.
457: \label{eq:tlc_n5_1}
458: \ee
459: This gives us the velocity of the reverse shock as,
460: \be
461: {v}_{r} ^\ast (t^\ast ) =
462: \frac{n-3}{n} \, \left[\frac{27}{4\pi}\frac{1}{n(n-3)
463: l^{2}\phi}\left\{\frac{10}{3}\left(\frac{n-5}{n-3}\right)\right\}^{\frac{n-3}
464: {2}}\right]
465: ^{1/n} \, \bigl (t^{\ast} \bigr ) ^{ -3/n} \,,
466: \ee
467: and correspondingly,
468: \bea
469: \tilde{v}_{r}^\ast (t^\ast )&=&\frac{R_{r}}{t} - v_{r}^\ast \nonumber\\
470: &=&
471: \frac{3}{n}\left[\frac{27}{4\pi}\frac{1}{n(n-3)l^{2}\phi}\left\{\frac{10}
472: {3}\left(\frac{n-5}{n-3}\right)\right\}^{\frac{n-3}{2}}\right]^{1/n}
473: t^{\ast -3/n} \,.
474: \label{eq:tlc_n5_2}
475: \eea
476: This is the result arrived at by Chevalier (1982) and Nadyozhin (1985) -
477: henceforth referred to as `the CN solution'. The transition of a supernova
478: remnant with $n > 5$ to this stage has been shown to occur extremely fast
479: (TM99).
480: Thus for all practical purposes,% as long as we do not talk of too early a time,
481: we may assume that the remnant begins with the CN stage.
482:
483: The reverse shock then moves into the core at $t = t_{core}$.
484: TM99 assumed
485: that the reverse shock motion after this time is also described by a
486: Sedov-Taylor trajectory - which leads to better match between the two regimes
487: in terms of continuity of the shock radius and velocity.
488: In the case of $n > 5$, therefore, if we replace $t_{ST}$ by $t_{core}$ in
489: the ST stage equations above, the same analysis follows through.
490: The numerical values of the parameters $t_{core}, \tilde{a}_{r,core}$ and
491: $\tilde{v}_{r,core}$ are, of course, different from the parameters for the
492: $n < 3$ case: viz $t_{ST}$ and $\tilde{a}_{r,ST}$.
493: We have,
494: \bea
495: R_r^\ast (t^\ast ) &=& t^\ast \Bigl [ \frac{R_{r,core}^\ast }{t_{core}^\ast }
496: - \tilde{a}_{r,core}^\ast (t^\ast -t_{core}^\ast ) \nonumber\\
497: && - (\tilde{v}_{r,core}^\ast
498: - \tilde{a}_{r,core}^\ast t_{core}^\ast ) ln\left(\frac{t^\ast }{t_{core}^\ast
499: }\right) \Bigr ] \nonumber\\
500: \tilde{v}_r^\ast (t^\ast ) &=& \tilde{v}_{r,core}^\ast + \tilde{a}_{r,core}
501: ^\ast (t^\ast - t_{core}^\ast ) \,.
502: \label{eq:tmc_n5_1}
503: \eea
504: The values of the parameters used in our calculations have been taken from TM99,
505: which have been checked with results from simulations.
506:
507:
508: \subsubsection{Summary of ED and ST solutions}
509: We thus have the final solutions for the reverse shock
510: as follows. In the case of $n < 3$,
511: equations (\ref{eq:ednl3_1}; first part) and (\ref{eq:ednl3_2}) describe the
512: solutions in the ED stage, and, thereafter, equations (\ref{eq:stnl3_1}) and
513: (\ref{eq:stnl3_2}) are appropriate for solutions in the ST stage.
514: For the case of $n>5$, solutions are given by
515: equations (\ref{eq:tlc_n5_1}) and (\ref{eq:tlc_n5_2})
516: in the $t < t_{core}$ stage. Afterwards,
517: in the $t > t_{core}$ stage, equations (\ref{eq:tmc_n5_1})
518: describe the solutions. The analytical solutions remain valid
519: until $t_\ast \sim 2.2$ (TM99), and we calculate the evolution
520: of shocks until this point (except for $n=4$; see below).
521:
522: For a given set of values of $E$ and $M_{ej}$, we can therefore
523: solve for reverse shock position and relative velocity, assuming a
524: density distribution described by index $n$. The parameter $\alpha$
525: in different cases of $n$, is given by,
526: \be
527: \alpha\equiv {E \over (1/2) M_{ej} v_{ej}^2}=\Bigl ( {3 -n \over 5-n} \Bigr )
528: \Bigl ( {w_{core}^{-(5-n)}-n/5 \over w_{core}^{-(3-n)}-n/3} \Bigr )
529: w_{core}^2 \,,
530: \ee
531: For $n<3$, we calculate this in the limit of $w_{core}=0$, and for
532: $n>5$ (including the case of $n=4$)
533: we assume (following TM99) a value of $w_{core}=0.1$. Values
534: of other parameters (such as $l, \phi$) are taken from appropriate tables
535: of TM99 for the relevant values of $n$.
536:
537:
538: We also consider the special case of $n=4$, for which the mass integral
539: diverges without a core. We follow the prescriptions of TM99 (their section 8)
540: for this case (and, again, assume $w_{core}=0.1$),
541: and track the reverse shock with only ED solution without
542: any transition to the ST case. According to TM99, based on their comparison
543: of analytical and numerical results, this prescription gives tolerably good
544: results (accurate to within $24\%$) until $t^\ast =1.2$.
545:
546: \subsection{Arrival time of reverse shock}
547: We have introduced a label to a given shell in the supernova ejecta by $w
548: \equiv w_f/l =\frac{R_r(t)/t}{v_{ej}}$, the ratio of the velocity of the ejecta
549: in the shell to the maximum ejecta velocity. The ejecta in a shell specified by
550: a given $w$ first undergoes free expansion (in the ED stage) and
551: self-similar expansion (in the ST stage) until hit by
552: the incoming reverse shock wave. This occurs at a time, which we call
553: $t_{hit}$, which is a function of $w$ and which is found by equating $w$ to
554: the $v_{r}/v_{ej}$
555: found from the position of the reverse shock. That is,
556: \be w = \frac{v_{r}}{v_{ej}} = \frac{R_{r}}{R_{ej}}
557: \Rightarrow R_{r} = w R_{ej} = w v_{ej} t_{hit} \,.
558: \ee
559: Or, in terms of dimensionless variables,
560: \be
561: R_{r}^\ast = w\cdot \frac{v_{ej}t_{hit}}{R_{ch}} = \left( w\cdot \frac{v_{ej}
562: t_{ch}}{R_{ch}} \right) t_{hit} ^\ast \,.
563: \ee
564: This expression for $R_{r}^\ast $ in terms of $t^\ast $ is equated to the
565: explicit solution for the reverse shock, when available, as a function of time,
566: and the resultant linear equation in $t_{hit}^{*}$ and $w$ is solved using
567: the bisection method for each given $w$. This is applicable to the steep
568: ($n > 5$) ejecta case (both the CN and ST stage) as well as to the ST stage
569: for shallow ($n < 3$) ejecta. For the ED stage of shallow ejecta, we use
570: $t_{hit}^{*}$\
571: as a function of $R_{r}^{*}$ in equation (\ref{eq:ednl3_1}) and then
572: employ the bisection method exactly as before to solve for $t^\ast _{hit}$.
573: Either way, we can determine $t^\ast _{hit}$ given any shell labeled
574: by a $w$ in the SN ejecta.
575:
576: \subsection{Density and temperature evolution in ejecta shells}
577: We assume that gas in a given shell of the ejecta evolves adiabatically before
578: $t_{hit}$. The temperature is then raised to the postshock temperature and the
579: density acquires a jump, corresponding
580: to the value of relative shock velocity $\tilde{v}_r$. After $t_{hit}$, the
581: gas again evolves adiabatically, that is, its density decreases as $t^{-3}$
582: (equation \ref{eq:den}). Its temperature decreases as $T \propto \rho^
583: {\gamma -1} \propto t^{-2}$, for $\gamma =5/3$ appropriate for monoatomic gas.
584: For a shell with a certain value of $w$, therefore, we have the following
585: evolution of density,
586: \be
587: \rho(t,w) = \left\{ \begin{array} {ll}
588: \frac{M_{ej}}{v_{ej}^3}
589: f(w)t^{-3}, & \mbox{$t < t_{hit}$} \\
590: 4 \frac{M_{ej}}{v_{ej}^3} f(w) t^{-3}
591: , & \mbox{$t_{hit} \le t < 2.2 t_{ch}$}
592: \end{array}
593: \right.
594: \label{eq:den_ev}
595: \ee
596: and temperature,
597: \be
598: T(t,w) = \left\{ \begin{array} {ll} T_i (\frac{t}{t_i})^{-2},
599: & \mbox{$t < t_{hit}$} \\
600: T_{rshock} (\frac{t}{t_{hit}})^{-2},
601: , & \mbox{$ t_{hit} \le t < 2.2 t_{ch}$}
602: \end{array}
603: \right.
604: \ee
605: where $T_{rshock}=\frac{3}{16} \mu m_p \tilde{v_r}^2/k_B$, the post-shock
606: temperature due to the reverse shock with relative velocity $\tilde{v_r}$.
607: For a completely ionized gas the mean molecular weight $\mu=0.6$.
608: We assume a value of $T_i=5400$ K at $t_i=70$ days, motivated by
609: photometric observations of SN1987A (Catchpole \etal 1987) where it was
610: estimated that the photospheric temperature at $\sim 70$ d after the
611: explosion was $\sim 5400$ K. We have also checked that our results do not
612: depend strongly on these assumptions.
613: %estimate for the temperature.% if the temperature is higher then the sputtering
614: %will also be more efficient than described below.
615:
616:
617: \subsection{Sputtering of dust}
618: Given the above mentioned evolution of density and temperature in shells
619: of ejecta labeled by different values of $w$, we can compute the steady
620: decrease in grain radius, for different types of dust grains. We use the
621: polynomial fit given by Tielens et al (1994) for graphites and silicates
622: as a function of gas temperature. They expressed the rate, $(1/n_H) {da \over
623: dt}$ in powers of $\log _{10} T$, with coefficients for different grain
624: composition tabulated in their
625: Table 4 (see their equation 4.21).
626: These rates are consistent with recent results of Nozawa \etal (2006).
627:
628: We follow a shell of ejecta material labeled by a
629: value $0 < w < 1$ from
630: $\sim$ 10 years after the explosion (by which time dust grains are
631: believed to have formed in the cooling ejecta, since the formation time
632: scale is of order a few years (see, e.g., Todini \& Ferrara 2001)),
633: to when the reverse shock hits the shell at $t_{hit}$ and beyond, until a
634: time $t^\ast \sim 2.2$, which is the limit of the validity of the analytical
635: solutions.
636:
637:
638: \section{Results}
639: We have calculated the effect of reverse shocks on sputtering of graphite
640: and silicate grains for typical value of explosion energy, stellar ejecta
641: mass and ambient density. Below we show the results for $E=10^{51}$ erg,
642: $M_{ej}=10^{34}$ g, $\rho_0=10^{-24}$ g cm$^{-3}$.
643:
644: First, we show the evolution of grain radius (for graphites and silicates) for
645: two dust grains embedded in two particular shells, for example, in ejecta
646: characterized by $n=0$ and $E=10^{51}$ erg. We show in Figure 1 the cases
647: for the shells marked $w=0.3$, and $0.6$. The reverse shock passes
648: through these two shells, first hitting the $w=0.6$ shell at $\sim 0.5
649: t_{ch} \sim 0.8 \times 10^3 n_0^{-1/3}$ yr, and then
650: the $w=0.3$ shell at $\sim 1.2
651: t_{ch} \sim 2 \times 10^3 n_0^{-1/3}$ yr. The bottom panel of Figure 1
652: shows
653: the rise in temperature for these two shells followed by adiabatic cooling.
654: Initially the temperature drops to very low temperatures due to strong
655: adiabatic cooling, although,
656: in reality, gas cannot be cooled below the temperature of the cosmic
657: microwave background, but we have not used such a lower bound in our
658: calculations. At any rate, the post-shock temperature does not depend on the
659: low temperature of the gas upstream, being determined only by the strength
660: of the reverse shock, and so the results of our calculation
661: remains realistic in spite of the strong cooling before the reverse shock hits
662: the gas in a given shell.
663:
664:
665: Figure 2 plots the ratio of final to initial grain sizes as a function
666: of the initial grain sizes for silicates and graphites forming (and being
667: sputtered) in shells $w=0.2, 0.4, 0.6$ for the particular case of $n=0$.
668: {\it Since the sputtering yield of silicate grains is larger than that of
669: graphites for a given set of parameters}, silicates are eroded more rapidly
670: than graphite grains for sputtering
671: in a given ejecta shell. Also, the relatively outer shell of $w=0.6$
672: experiences less sputtering than the inner shells of $w=0.2$ and $0.4$. This
673: is expected since the reverse shock (relative) speed picks up as it plows
674: through the ejecta, increasing with decreasing values of $w$.
675: To illustrate this point,
676: we plot the trend of the relative velocity of the reverse shock as a function
677: of $w$ for the shells hit by the shock in Figure 3, for $n=0$ (solid),
678: $n=2$ (dotted), and $n=6$ (dashed). % and $n=8$ (dash-and-dotted).
679: The relative
680: velocity is a rising function with decreasing $w$, for $n=0$ and
681: for $n>5$, and it is a peaked function for $n=2$ (and also for $n=4$, which
682: is not shown here, but as shown by TM99).
683:
684: However, we notice in Figure 2 that grains (both graphites and silicates)
685: in shell $w=0.2$ (solid curve) are sputtered to a somewhat
686: lesser extent than grains in the $w=0.4$ shell, although the relative speed
687: of the reverse shock should be (according to Figure 3, solid curve for $n=0$)
688: higher when it hits the $w=0.2$ shell than at the $w=0.4$ shell.
689: Although the post-shock temperature at $w=0.2$ is higher than at $w=0.4$,
690: the shells in the inner region
691: encounter the reverse shock later
692: than shells in the outer region of ejecta,
693: and by the time they encounter it, the density in the inner shells
694: would have decreased rapidly (see equation \ref{eq:den_ev}). The rarefaction
695: in the inner shells at the time of encounter with the reverse shocks
696: lessens the effect of dust sputtering in these shells.
697:
698: We illustrate this point by considering grains which are sputtered below
699: a certain size, say, $10^{-7}$ cm ($10^{-3} \, \mu$m),
700: as a function of the shell in which
701: they reside (and get sputtered).
702: We determine the
703: maximum size of grains which are sputtered below a size $\sim 10^{-7}$ cm,
704: as a function of $w$, for different shells, and present the results in
705: Figure 4.
706: Curves for different
707: values of $n$ show the maximum size of grains sputtered below $0.001 \, \mu$m
708: as a function of the parameter $w$ (left panels), as well as against
709: the fraction of ejecta mass that is
710: contained inside of the concerned shell, $M(<w)/M_{ej}$ in the right panels.
711: Results for
712: graphites and silicate grains are shown in the upper and lower
713: panels respectively.
714:
715:
716: Figure 4 clearly shows that sputtering becomes intense in intermediate
717: shells, and decreases (as a result of rarefaction) with decreasing value
718: of $w$ for small values of $w$.
719: The sizes of grains
720: which are sputtered below $10 ^{-3} \mu $m
721: lie in the range of $\le 10^{-3} \, \mu$m
722: $\hbox{--}10^{-2.6} \, \mu$m, with silicates more
723: strongly sputtered than graphites. In all cases,
724: the curves of
725: the maximum
726: size of sputtered grains
727: increases with decreasing values of $w$ and $M(<w)/M_{ej}$, finally
728: plunging to very low values for $w \ll 1$,
729: indicating negligible effect of the reverse shock deep
730: inside the ejecta.
731: For example, the (dotted)
732: curve for $n=2$ shows a peak at $w \sim 0.17$ below which the maximum
733: size of sputtered grains decreases.
734:
735:
736: Also, although it is not explicitly shown
737: in Figure 4, the corresponding results for $n >5$ show that reverse shocks
738: do not penetrate most of the ejecta mass in the time period of the
739: validity range
740: of analytical solutions $t_\ast \sim 2.2$.
741: Analytical means allow us to
742: probe deep into the ejecta mass distribution only for $n < 3$, and
743: full hydrodynamical models are needed to address the cases of steeper
744: profiles. For steep
745: density distributions, the limiting
746: time of $t=2.2t_{ch}$ is reached by the time the reverse shock only skims
747: the surface of the ejecta, as far as mass fraction is concerned. The
748: characteristic time in the case considered here ($E=10^{51}$ erg, $M_{ej}=
749: 10^{34}$ g, ambient density $\rho_0=10^{-24}$ g cm$^{-3}$)
750: is $t_{ch} \sim 1.7 \times 10^3$ yr. Although, by the limiting
751: time of $2.2 t_{ch} \sim 3.7 \times 10^3$ yr for the validity of self-similar
752: regime, reverse shocks can hit shells with small $w$ in these steep cases,
753: the mass contained inside of these shells remains very large.
754: In other words, in
755: these cases we cannot probe what happens to most of the ejecta mass, or the
756: dust contained in them with analytical means, within the limits of
757: the self-similar regime of shock evolution.
758:
759:
760: We also note that a substantial amount of grain sputtering takes place
761: in the hot gas between the forward and reverse shock that analytical
762: methods cannot capture. Detailed hydrodynamical calculations
763: are needed to address these issues, and our result can only provide a
764: lower limit to the total mass fraction of dust that is destroyed.
765:
766:
767: Finally, we calculate the total dust mass sputtered away as a function of
768: initial grain sizes in the following manner. We first calculate for a given
769: shell $w$, the final grain size $a_f (a_i,w)$ corresponding to the
770: initial grain
771: size $a_i$. Then the fraction $1-(a_f/a_i)^3$ is the total dust mass sputtered
772: for grains formed in this shell. We then calculate the total dust mass
773: sputtered for all shells (for a given initial grain size $a_i$),
774: weighted by the shell mass in which they are situated, as given by,
775: \be
776: f_d (a_i)= \int \bigl [ 1-({a_f \over a_i})^3 \bigr ] \, {d M(<w) \over
777: M_{ej}} \,,
778: \ee
779: where $M(<w)$ is the ejecta mass contained within the shell $w$.
780:
781:
782: We plot these weighted fractions $f_d (a_i)$ for graphite and silicate grains
783: as functions of $a_i$ for $n=0,2,4$ in Figure 5. Thick curves denote the
784: results for silicates and thin ones are for graphite grains. As expected,
785: the sputtered fraction decreases with increasing $a_i$, and the fraction
786: for silicates is in general larger than that for graphites.
787:
788: The total dust mass fraction that is destroyed can then be evaluated from these
789: results by convolving $f_d (a_i)$ with an initial grain size distribution.
790: As mentioned earlier, Nozawa \etal (2003) found that the size distribution
791: of synthesized grains is bounded by two power laws, with indices
792: $-3.5$ for larger
793: grains and $-2.5$ for smaller grains. We therefore
794: estimate the total dust mass fraction destroyed, assuming a size
795: distribution $n(a) da \propto a^{-p} da$, with $p=2.5$ and $3.5$.
796: The destroyed fraction of dust mass is given by,
797: \be
798: f={\int n(a) a^3 f_d(a) da \over \int n(a) a^3 da} \\,
799: \ee
800: where the grain volume scales as $a^3$. The results for graphites
801: and silicates are presented in Table 1, for $n=0,2,4$
802: ($E=10^{51}$ erg, $M_{ej}=10^{34}$ g), leaving the cases
803: of $n > 5$ for which the reverse shocks do not reach the interiors of the
804: ejecta within the validity range of self-similar solutions.
805: We assume the lower and upper limit for the integral to be $10^{-7}$ cm
806: and $3 \times
807: 10^{-5}$ cm respectively, as explained earlier in \S 2. We also
808: calculate the destroyed fractions for a truncated grain size distribution,
809: with an upper limit of $10^{-6}$ cm, and the results are shown inside brackets
810: in Table 1 for comparison. %The results
811: %do not depend strongly on these assumptions as long as the integration
812: %includes the range of $10^{-7}\hbox{--}10^{-6}$cm that contains most of
813: %the dust mass and above which sputtering is negligible.
814:
815: We have further studied the variation of the destroyed dust mass fraction
816: with explosion energy and ejected mass. Tables 2 and 3 show the fraction
817: of dust mass supttered for two sets of parameters: Table 2 for $M_{ej}=10^{34}$
818: g and $E=10^{52}$ erg, and Table 3 for $M_{ej}=2 \times 10^{34}$ g and
819: $E=10^{51}$ erg. Figure 6 plots the varition of
820: (logarithm) of the destroyed dust mass fraction, $f$, for silicates, for
821: different values of explosion energy and ejected mass, and for different
822: mass profiles ($n=0,2,4$). In the upper panel, we plot the variation of
823: $f$ with $M_{ej}$ (in logarithm), keeping the explosion energy constant
824: at $E=10^{51}$ erg, and in the lower palnel, we plot (logarithms of) $f$
825: with explosion energy $E$, keeping the ejected mass fixed at $M_{ej}10^{34}$
826: g. The destroyed fraction generally increases with ejected mass,
827: but its variation with explosion energy is not monotonic. The reason is that
828: the process of dust destruction depends mainly on two factors: the speed of
829: the reverse shock (which depends on the characteristic speed $v_{ch}$) and the
830: characteristic time, $t_{ch}$. Now, $v_{ch} \propto M_{ej}^{-1/2} E^{1/2}$,
831: whereas $t_{ch} \propto M_{ej}^{5/6} E^{-1/2}$. It would appear that a
832: change in ejected mass manifests itself in increasing $t_{ch}$, and thereby
833: increasing the destroyed fraction, whereas, a change in $E$ affects $t_{ch}$
834: and $v_{ch}$ in a complicated manner, which results in the variation shown in
835: the bottom panel of Figure 6.
836:
837:
838: \section{Discussion}
839: We therefore find that a mass fraction
840: $ \leq 20$ \% of silicates and %$\le 5$\% of
841: graphite grains
842: created in cooling
843: ejecta can be sputtered away by the reverse shock if the
844: density distribution of the ejecta has a shalow profile ($n < 5$).
845: Nozawa et al (2007) recently studied sputtering of dust in reverse shock
846: and found a destroyed
847: dust mass fraction of $0.2\hbox{--}1$; their result for
848: an ambient hydrogen
849: number density
850: of $1$ cm$^{-3}$ and a progenitor mass of $13$ M$_{\odot}$ is $\sim 0.7$
851: (their Table 1). In fact, their result for the `unmixed grains' model
852: (which has to do with spatial mixing of the core heavy-element
853: distribution)
854: in which the maximum size is $0.01 \, \mu$m
855: is $\sim 0.5$.
856: It should be noted, however, that a considerable amount
857: of sputtering in their calculation is caused by the hot plasma between
858: the forward and the reverse shocks (see \S 3.3 in Nozawa et al (2007)),
859: which we have not considered. In
860: our case, for analytical simplicity, we have assumed the post-reverse shocked
861: gas to be cooling adiabatically.
862: %Also, Nozawa \etal (2007) have considered other destruction channels
863: %including non-thermal sputtering arising from grain gyro-motion and
864: %grain-grain collision, and which are beyond the scope of the present
865: %paper as they require assumptions about betatron acceleration of the
866: %Larmor orbits of charged grains.
867: Our results are, therefore, expected
868: to give lower bounds on the actual dust mass fraction destroyed
869: in SN remnants.
870:
871: There is another reason why the destroyed fraction of dust mass
872: calculated here is lower than that of Nozawa et al (2007) and Bianchi
873: \& Schneider (2007). Strictly speaking,
874: the destruction rate of Tielens et al (1994) used
875: here is valid for a gas with solar abundance. The gas in the dust
876: formation region in the supernova ejecta is expected to be metal
877: rich, and the destruction of dust grains in this gas can be more
878: efficient in this case, since sputtering yields by metal ions are
879: much higher in general than by hydrogen and helium ions.
880:
881:
882: It should be noted that our results for the destroyed fraction is
883: independent of the ambient density, since the gas density enters
884: into our calculation only to change $t_{ch}$ (notice in equations
885: \ref{eq:char} that $v_{ch}$ is independent of ambient density, and so
886: the relative speed of reverse shock, and in turn the post-shock temperature,
887: are also independent of ambient density). In reality, however, a lot
888: of grain sputtering would take place between the forward and reverse shock,
889: where the gas density will crucially depend on the ambient density. This
890: aspect of grain sputtering is neglected in our analytical calculation,
891: but is captured in the simulation results of Bianchi \& Schneider (2007)
892: and Nozawa \etal (2007).
893:
894: Another difference between our results and those of Bianchi \& Schneider
895: (2007) arises from the assumption of initial grain size distribution.
896: Bianchi \& Schneider (2007) used grains of small size in their
897: calculation, with silicate grains smaller than $\le 0.01 \, \mu$m and
898: graphites smaller than $\le 0.05 \, \mu$m. Also, their size distribution
899: is more confined to a narrow range than the power law distribution assumed
900: here. Both these factors would enhance the destroyed fraction of dust
901: mass in reverse shocks. For example, if the sizes of $SiO_2$ grains have
902: a delta-function distribution around $a_i \sim 25 \, \AA$ (see Figure 1
903: of Bianchi \& Schneider 2007), then the curves in Figure 2 of
904: the present paper (say, for $w=0.2$) yield a mass destruction fraction of
905: $\sim 22$ \% (with $a_f/a_i \sim 0.92$). For $M_gSiO_3$ grains of size
906: $\sim 45 \, \AA$, we similarly estimate a destroyed mass fraction of
907: $\sim 14$ \%. From Figure 2, we estimate that one requires $a_I\sim 10 \, \AA$
908: in order to get $\sim 50$\% destruction of silicate grains.
909:
910:
911: A note on the observed density distribution of supernovae ejecta is in
912: order here.
913: Chevalier \& Fransson (1994) studied a model of the ejecta
914: from a Type II supernova with a shallow ($n<3$) core surrounded by a
915: steep ($n>5$) envelope. In the case of Type Ia supernovae, a steep density
916: distribution in the ejecta mass has also been discussed in the literature
917: (Colgate \& McKee (1969); Dwarakadas \& Chevalier (1997)).
918: There has been success, however, in modelling
919: data of Type Ia supernova with uniformly distributed ($n\sim0$) ejecta
920: (Hamilton \& Sarazin (1984); Hamilton, Sarazin \& Szymkowiak (1986)).
921:
922:
923: It is worth considering what observations might improve our
924: understanding of the net dust production by SNe. Far-infrared and
925: sub-millimeter observations of reverse-shocked dust in young SNe
926: by {\it Spitzer}, {\it Herschel}, and other facilities will help,
927: but our theoretical studies suggest that unshocked (cold) dust
928: may harbor a considerable mass of undetected dust. To constrain
929: this dust and distinguish it from surrounding interstellar clouds
930: will probably require sub-mm observations with small beam sizes.
931: This brings in complications of the circumstellar environment of
932: young SNe. Furthermore, the long times (thousands of years) required
933: for reverse shocks to reach the core ejecta suggest that late-time
934: observations of supernova remnants would be useful.
935:
936:
937: \section{Summary}
938: We have studied the effect of reverse shocks analytically,
939: in the regime of self-similar
940: evolution, on the sputtering of
941: dust grains that are believed to form in cooling ejecta. For representative
942: cases we found that fractions of dust mass that is destroyed are of order
943: of order $1\hbox{--}20$ \% of silicates and %$\le 5$\% for
944: %$15\hbox{--}20$ \% silicates and $5\hbox{--}10$\% for
945: graphites,
946: for SN of explosion energy $ \sim 10^{51}$ erg, ejecta mass
947: $\sim 10^{34}$ g, and an ambient density $ \sim 10^{-24}$ g cm$^{-3}$.
948: Our analytical formalism provides only a lower bound on the dust mass
949: fraction that is destroyed by the reverse shock,
950: since it ignores further sputtering of grains
951: in hot plasma between the forward and reverse shocks. Our results are,
952: therefore,
953: consistent
954: with the recent estimates from numerical simulations which
955: include these additional effects.
956: Furthermore, Our study provides a formalism to generalize these results
957: to cases with different values of SN parameters.
958:
959:
960:
961:
962: \acknowledgements We thank Drs. R. Chevalier, E. Dwek, T. Nozawa, A. Ray
963: and Y. Shchekinov for helpful discussions.
964: This work was supported at the University of Colorado by NASA Theory
965: grant NNX07-AG77G and NSF theory grant AST07-07474.
966:
967: \clearpage
968:
969: \begin{thebibliography}{}
970:
971:
972: \bibitem[]{bertoldi} Bertoldi, F., Carilli, C. L., Cox, P., Fan, X.,
973: Strauss, M. A., Beelen, A., Omont, A. \& Zylka, R. 2003, A\&A, 406, L55
974: \bibitem[]{bianchi} Bianchi, S. \& Schneider, R. 2007, MNRAS, 378, 973
975: \bibitem[]{catch} Catchpole, R. M. et al., MNRAS, 229, 15
976: \bibitem[]{chev1} Chevalier, R. A. 1982, ApJ, 258, 790
977: \bibitem[]{colgate} Colgate, S. A. \& McKee, C. 1969, 157, 623
978: \bibitem[]{deneault} Deneault, E. A., Clayton, D. D. \& Heger, A. 2003, ApJ,
979: 594, 312
980: \bibitem[]{dunne} Dunne, L., Eales, S., Ivison, R., Morgan, H., Edmunds, M.
981: 2003, Nature, 424, 285
982: \bibitem[]{dwek2} Dwek, E. 2006, Science, 313, 178
983: \bibitem[]{dwek3} Dwek, E., \& Arendt, R. G. 2007, in AIP Conf. Proc.,
984: Supernova 1987A: 20 Years After, ed. S. Immler,
985: K. Weiler, R. McCray, Vol. 937, 58
986: % 2007, {\it Dust-Gas Interaction
987: %in SN~1987A}, in AIP Proceedings,
988: % ``Supernova 1987A: 20 Years After" (astro-ph/0705.3796)
989: \bibitem[]{dwek4} Dwek, E., \& Werner, M. W. 1981, ApJ, 248, 138
990: \bibitem[]{gerardy} Gerardy, C. L., Fessen, R. A., H\"offlich, P. \& Wheeler,
991: J. C. 2000, AJ, 119, 2968
992: \bibitem[]{gerardy} Gerardy, C. L. \etal 2002, ApJ, 575, 1007
993: \bibitem[]{ham2} Hamilton, A. J. S. \& Sarazin, C. L. 1984, ApJ, 281, 682
994: \bibitem[]{ham3} Hamilton, A. J. S., Sarazin, C. L. \& Szymkowiak, A. E.
995: 1986, ApJ, 300, 698
996: \bibitem[]{kozasa} Kozasa, T., Hasegawa, H. \& Nomoto, K. 1989, ApJ, 344, 325
997: \bibitem[]{krause} Krause, O., et al.\ 2004, Nature, 432, 596
998: \bibitem[]{ledoux} Ledoux, C., Bergeron, J., Petitjean, P. 2002, A\&A, 385,
999: 802
1000: \bibitem[]{mcray} McCray, R. 1993, ARA\&A, 31, 175
1001: \bibitem[]{mcray2} McCray, R. 2007, in AIP Conf. Proc.,
1002: Supernova 1987A: 20 Years After, ed. S. Immler,
1003: K. Weiler, R. McCray, Vol. 937, 3
1004: %{\it Supernova 1987A at Age 20},
1005: %in AIP Proceedings,
1006: % ``Supernova 1987A: 20 Years After" (astro-ph/0705.3796).
1007: \bibitem[]{meikle} Meikle, W.P.S., Mattila, S., et al.\ 2007, ApJ, 665, 608
1008: \bibitem[]{morgan} Morgan, H. L. \& Edmunds, M. G. 2003, MNRAS, 343, 427
1009: \bibitem[]{moseley} Moseley, S. H., Dwek, E., Giaccum, W., Graham, J. R.,
1010: Loewenstein, R. F., Silverberg, R. F. 1989, Nature, 340, 697
1011: \bibitem[]{nadyozhin} Nadyozhin, D. K. 1985, Ap\&SS, 112, 225
1012: \bibitem[]{nozawa} Nozawa, T., Kozasa, T., Umeda, H., Maeda, K. \& Nomoto,
1013: K. 2003, ApJ, 598, 785
1014: \bibitem[]{noz} Nozawa, T., Kozasa, T., Habe, A. 2006, ApJ, 648, 435
1015: \bibitem[]{nozawa2} Nozawa, T. et al.\ 2007, astro-ph/0706.0383
1016: \bibitem[]{pettini1} Pettini, M., Smith, L. J., Hunstead, R. W. \&
1017: King, D. L. 1994, ApJ, 426, 79
1018: \bibitem[]{schneider} Schneider, R., Ferrara, A. \& Salvaterra, R. 2004,
1019: MNRAS, 351, 1379
1020: \bibitem[]{Seab83} Seab, C. G. \& Shull, J. M. 1983, ApJ, 275, 652
1021: \bibitem[]{shull} Shull, J. M. 1978, ApJ, 226, 858
1022: \bibitem[]{stani05} Stanimirovic, S., et al. 2005, ApJ, 632, L103
1023: \bibitem[]{sugerman06} Sugerman, B. E. K. et al. 2006, Science, 313, 196
1024: \bibitem[]{tielens} Tielens, A. G. G. M., McKee, C. F., Seab, C. G. \&
1025: Hollenbach, D. J., 1994, ApJ, 431, 321
1026: \bibitem[]{todini} Todini, P. \& Ferrara, A. 2001, MNRAS, 325, 726
1027: \bibitem[]{TM99} Truelove, J. K. \& McKee, C. F. 1999, ApJ, 120, 299 (TM99)
1028: \bibitem[]{whittet} Whittet, D. C. B. 1992, {\it Dust in the Galactic
1029: Environment}, IOP Publishing, Great Britain.
1030: %\bibitem[]{williams} Williams, B. J., Borkowski, K., et al.\ 2006, ApJ, 652, L33
1031:
1032: \end{thebibliography}
1033:
1034: \clearpage
1035:
1036: \begin{table}
1037: \centerline{Table 1 : Fraction of dust mass sputtered for sizes ranging from
1038: $10^{-4} \, \mu$m to $a_{max}=0.3 \, \mu$m (or $a_{max}=0.01 \, \mu$m)
1039: }
1040: \vskip 0.2in
1041: \begin{center}
1042: \begin{tabular}{|c||c||c|c|}
1043: \hline
1044: grain ($dn/da$) & $n=0 $ & $n=2$ & $n=4$ \\
1045: \hline
1046: \hline
1047: Graphite ($\propto a^{-3.5}$) & $0.004 \, (0.02) $ & $0.01 \,(0.05) $ & $0.007 \,
1048: (0.04)$ \\
1049: \hline
1050: ($\propto a^{-2.5}$) & $6 \times 10^{-4} \,(0.01)$ & $0.001 \,(0.03) $ & $0.001 \,(0.03) $ \\
1051: \hline
1052: Silicate ($\propto a^{-3.5}$)& $0.01 \,(0.07) $ & $0.03 \, (0.13) $ & $0.02 \,(0.10)$ \\
1053: \hline
1054: ($\propto a^{-2.5}$) & $0.002 \, (0.05) $ & $0.005 \, (0.10) $ & $0.004 \, (0.07)$ \\
1055: \hline\hline
1056: \end{tabular}
1057: \end{center}
1058: \end{table}
1059:
1060:
1061: \begin{table}
1062: \centerline{Table 2 : Fraction of dust mass sputtered for $M_{ej}= 10^{34}$
1063: g, $E_{ej}=10^{52}$ erg.
1064: % for sizes ranging from
1065: %$10^{-4} \, \mu$m to $a_{max}=0.3 \, \mu$m (or $a_{max}=0.01 \, \mu$m)
1066: }
1067: \vskip 0.2in
1068: \begin{center}
1069: \begin{tabular}{|c||c||c|c|}
1070: \hline
1071: grain ($dn/da$) & $n=0 $ & $n=2$ & $n=4$ \\
1072: \hline
1073: \hline
1074: Graphite ($\propto a^{-3.5}$) & $0.001 \, (0.008) $ & $0.003 \,(0.01) $ & $0.005 \,
1075: (0.03)$ \\
1076: \hline
1077: ($\propto a^{-2.5}$) & $2 \times 10^{-4} \,(0.005)$ & $4 \times 10^{-4} \,(0.01) $ & $7 \times 10^{-4} \,(0.02) $ \\
1078: \hline
1079: Silicate ($\propto a^{-3.5}$)& $0.006 \,(0.03) $ & $0.01 \, (0.06) $ & $0.05 \,(0.16)$ \\
1080: \hline
1081: ($\propto a^{-2.5}$) & $0.001 \, (0.02) $ & $0.002 \, (0.04) $ & $0.018 \, (0.14)$ \\
1082: \hline\hline
1083: \end{tabular}
1084: \end{center}
1085: \end{table}
1086:
1087: \begin{table}
1088: \centerline{Table 3 : Fraction of dust mass sputtered for $M_{ej}=
1089: 2 \times 10^{34}$ g,
1090: $E_{ej}=10^{51}$ erg.
1091: % for sizes ranging from
1092: %$10^{-4} \, \mu$m to $a_{max}=0.3 \, \mu$m (or $a_{max}=0.01 \, \mu$m)
1093: }
1094: \vskip 0.2in
1095: \begin{center}
1096: \begin{tabular}{|c||c||c|c|}
1097: \hline
1098: grain ($dn/da$) & $n=0 $ & $n=2$ & $n=4$ \\
1099: \hline
1100: \hline
1101: Graphite ($\propto a^{-3.5}$) & $0.006 \, (0.035) $ & $0.015 \,(0.08) $ & $0.01 \,
1102: (0.06)$ \\
1103: \hline
1104: ($\propto a^{-2.5}$) & $10^{-3} \,(0.02)$ & $0.003 \,(0.06) $ & $0.002 \,(0.04) $ \\
1105: \hline
1106: Silicate ($\propto a^{-3.5}$)& $0.02 \,(0.1) $ & $0.04 \, (0.18) $ & $0.03 \,(0.13)$ \\
1107: \hline
1108: ($\propto a^{-2.5}$) & $0.003 \, (0.07) $ & $0.008 \, (0.14) $ & $0.005 \, (0.10)$ \\
1109: \hline\hline
1110: \end{tabular}
1111: \end{center}
1112: \end{table}
1113:
1114:
1115:
1116:
1117: \clearpage
1118:
1119: \begin{figure}
1120: \epsscale{1.0}
1121: \plotone{f1.eps}
1122: \caption{
1123: The dust grain size as a function of time (where $t_\ast=t/t_{ch}$) is plotted
1124: for $n=0$, $E=10^{51}$ erg, $M_{ej}=10^{34}$ g
1125: in the upper panel, for two shells, with $w=0.6$
1126: and $w=0.3$. The shell $w=0.3$ is inside of $w=0.6$ shell, and is hit by the
1127: reverse shock at a later time. The initial grain size (solid line for graphites,
1128: and dotted, for silicates) is $1.3 \times 10^{-3}$ $\mu m$.
1129: The bottom panel plots gas temperature as a function of $t^\ast$ showing
1130: the rise at the instant of the reverse shock meeting the two particular shells.
1131: ($t_{ch}\sim 1.7 \times 10^3 n_0^{-1/3}$ yr.)
1132: }
1133: \end{figure}
1134:
1135: \clearpage
1136:
1137: \begin{figure}
1138: \epsscale{1.0}
1139: \plotone{f2.eps}
1140: \caption{
1141: Ratio of final to initial grain radius is plotted against initial grain
1142: radius (in $\mu$m) for silicates (thick lines) and graphites (thin lines),
1143: for grains in shells $w=0.2$ (solid),
1144: $w=0.4$ (dotted), $w=0.6$ (dashed) for the case
1145: of $n=0$.
1146: }
1147: \end{figure}
1148:
1149: \clearpage
1150:
1151: \begin{figure}
1152: \epsscale{1.0}
1153: \plotone{f3.eps}
1154: \caption{
1155: The dimensionless velocity of reverse shock in the frame of unshocked ejecta,
1156: $\tilde{v_r} ^\ast$
1157: (in the units of the characteristic velocity $v_{ch}$)
1158: is plotted against $w$ for $n=0,2,6$ (solid, dotted,
1159: dashed lines). For reference, the value of $v_{ch} \sim 3162$ km
1160: s$^{-1}$, for $E=10^{51}$ erg and $M_{ej}=10^{34}$ g.
1161: }
1162: \end{figure}
1163:
1164: \clearpage
1165:
1166: \begin{figure}
1167: \epsscale{1.0}
1168: \plotone{f4.eps}
1169: \caption{
1170: The values of maximum size, $\log a_{max}$ (in $\mu$m),
1171: below which grains are sputtered
1172: below $10^{-7}$ cm
1173: are shown against the parameter $w$ in the left panels, and against
1174: the fraction of ejecta mass inside
1175: the corresponding shell, $M(<w)/M_{ej}$, in the right panels.
1176: ($w=R(t)/(v_{ej}t)$, or equivalently, it
1177: is the ratio of the velocity of the ejecta in the shell to the
1178: maximum ejecta velocity.) The
1179: case of graphites are shown in upper panels and the case of silicates,
1180: in the lower panels. Dust
1181: is introduced at 10 yr, where the tracks begin. Solid, dotted, and dashed
1182: lines refer to the cases $n=0,2,4$ respectively.
1183: The energy of explosion is assumed to be $E_{ej}=10^{51}$ erg and ejecta
1184: mass is assumed to be $10^{34}$ g.
1185: }
1186: \end{figure}
1187:
1188: \clearpage
1189:
1190: \begin{figure}
1191: \epsscale{1.0}
1192: \plotone{f5.eps}
1193: \caption{
1194: Fraction of dust mass sputtered as a function of initial grain size,
1195: weighted by the mass of the shell in which they are formed,
1196: for $n=0$ (solid), $2$ (dotted) and $4$ (dashed), for silicates
1197: (thick curves) and graphites (thin curves).
1198: }
1199: \end{figure}
1200:
1201: \clearpage
1202:
1203: \begin{figure}
1204: \epsscale{1.0}
1205: \plotone{f6.eps}
1206: \caption{
1207: Fraction of dust mass sputtered is plotted in the upper
1208: panel as a function of mass of ejecta, keeping the explosion energy
1209: constant at $10^{51}$ erg, for different values of $n$. In the lower
1210: panel, the fraction of destroyed dust mass is plotted (for different $n$)
1211: as a function of explosion energy $E$, keeping the ejecta mass
1212: constant at $10^{34}$ g.
1213: }
1214: \end{figure}
1215:
1216: \clearpage
1217:
1218:
1219:
1220: \end{document}
1221:
1222: