Post on 05-Oct-2020
Pulsar glitches and neutron star masses
Alessandro Montoli
First-year PhD student workshop 2018
Milano, 10th October 2018
Neutron stars and pulsars
Extreme features of neutron stars:
• R ≈ 10 km
• M ≈ 1− 2M
• ρc & 1014 g / cm3
• P ≈ 10−3 − 101 s
• B ≈ 108 − 1015 G
Lighthouse model:
• Magnetic field axis misaligned with respectto the rotational axis
• Conversion of rotational energy intoelectromagnetic radiation
• Pulse frequency = rotational frequency
Pulsar glitches
-400 -200 0 200 400Days from Glitch
-374
9-3
744
ν (1
0-13 H
z s-1
)
•-400 -200 0 200 400
Days from Glitch
-374
7-3
742
00.
20.
40.
6 B0531+2151740.656
Δν
(μHz
)(a)
-400 -200 0 200 400
00.
6
B0531+2152084.072
(b)
-400 -200 0 200 400Days from Glitch
-376
0-3
740
-400-200 0 200 400
-40
48 B0531+21
53067.0780(c)
-400 -200 0 200 400Days from Glitch
-373
4-3
726 -400-200 0 200 400
00.
20.
40.
60.
8 B0531+2153970.1900
(d)
Four glitches of the Crab pulsar, Espinoza et al. (2011)
Two components: a normal charged one (visible) and a superfluid one.
Long recoveries ⇒ Effect due to superfluid components.
Diverse phenomenology:
• Mostly radiopulsars, but also magnetars and millisecond.• Periodic glitchers vs single glitchers.• Similar size vs. different size.
Neutron star structure
Glitch mechanism
• The normal component loses angular momentum.
• The superfluid rotates by forming an array ofquantised vortices.
• These vortices may pin to impurities of the crust.
The relative motion between the vortices and thesuperfluid itself causes a Magnus force. If the Magnusforce is strong enough to overcome the pinning force,vortices detach and a glitch occurs.
F
v
Critical lag
10-5
10-4
10-3
10-2
10-1
100
0 0.2 0.4 0.6 0.8 1
ωcr
[rad/
s]
x / Rd
M=1.1 MSunM=1.4 MSunM=1.9 MSun
∫vortex
fP dl =
∫vortex
fM dl
Critical lag for vortices extended to the whole star (Antonelli & Pizzochero,2017)
Maximal glitch
Angular momentum of the star:
L = IΩp + ∆L[Ωnp]
Time derivative of the angular momentum:
I Ωp + ∆L[Ωnp] = −I |Ω∞|
Integration between a time before and one after the glitch:
⇒ ∆Ωmax(t) =∆L[Ωnp(t)]
I
In the Newtonian framework:
∆L[Ωnp] =
∫d3x x2ρn(r) Ωnp
Lag as a function of time
Our aim is that of calculating the time dependence of the lag.
The equations that allow us to calculate this value are complicated and theydepend on the rotational parameters of the star (see Antonelli & Pizzochero2017).
We can use a simplified (and unified)prescription that allow us to study thequantitative trend of the lag:
Ωvp(x , ω∗) = min [Ωcrvp, ω
∗],
where Ωvp ≡ (1− εn)Ωnp and ω∗ ≡ |Ω∞|t.
Pulsar activity and ω∗act
The absolute pulsar activity is defined as:
Aa =1
Tobs
N∑i=1
∆Ωi .
We can also define a timescale for a pulsar that defines the mean waiting timefor a fictitious pulsar that exhibits only glitches of size ∆Ωobs and has activityAa:
tact =∆Ωobs
Aa.
Finally, we can find the nominal lag associated to this timescale:
ω∗act = tact|Ω∞|.
Mass estimate Mact
The maximal glitch ∆Ω(ω∗,M) depends only on the mass once themicrophysical parameters are fixed.
Measuring a maximum glitch ∆Ωobs and calculating the activity, we canestimate a mass for the pulsar inverting the relation:
∆Ω(ω∗act,Mact) = ∆Ωobs.
This mass estimate depends on the extension of the superfluid reservoir.
Critical lag
0.0 0.2 0.4 0.6 0.8 1.0r/R
10 4
10 3
10 2
10 1
100
101
Criti
cal l
ag (r
ad/s
)
Whole star1 n0
0.75 n00.68 n0
0.6 n0Crust
The critical lag depends on the cutoff considered.
Mass dependence on cuts
0.8
1.0
1.2
1.4
1.6
1.8
2.0
2.2
Mac
t (M
sun)
CrustWhole star
0.60 n0Whole star
0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0Largest observed glitch (1e-4 rad/s)
0.8
1.0
1.2
1.4
1.6
1.8
2.0
2.2
Mac
t (M
sun)
0.68 n0Whole star
0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0Largest observed glitch (1e-4 rad/s)
0.75 n0Whole star
0.8
0.9
1.0
1.1
1.2
1.3
1.4
1.5
1.6
Mas
s [M
sun]
J0537-6910J0537-6910J0537-6910 J0835-4510J0835-4510J0835-4510
0.45 0.50 0.55 0.60 0.65 0.70 0.75Cutoff [n0]
0.8
1.0
1.2
1.4
1.6
1.8
2.0
2.2
2.4
Mas
s [M
sun]
J1341-6220J1341-6220J1341-62200.45 0.50 0.55 0.60 0.65 0.70 0.75
Cutoff [n0]
J1801-2304J1801-2304J1801-2304
BSk20SLy4DDME2
Montoli, Antonelli & Pizzochero, arXiv:1809.07834Bayesian two-sample test: Antonelli, Gherardi, Montoli & Pizzochero, work in
progress.
Conclusions
• The mass estimate Mact depends on the extension of the superfluid region.
• Mact is generally lower with a smaller superfluid reservoir.
• Like in other models (Andersson et al. 2012, Chamel 2013, Delsate et al.2016), also here there is too little angular momentum stored in the crustfor glitches.
• A little bit more of superfluid reservoir is enough to describe glitches.
Open question: Where actually resides the superfluid involved in the glitch?
Thank you!
Mass upper limit Mmax
Maximum glitch = maximal glitch in the case of the critical lag.
∆Ωmax =π2
Iκ
∫ Rd
0dr r3 fP(r).
As for Mact, we can find a mass corresponding to the maximum glitch, Mmax.This is analytically independent on the extent of the superfluid region - as longas this extends at least in the pinning region - and entrainment paramters.
Already discussed in Pizzochero,Antonelli, Haskell & Seveso (2017) forthe Newtonian case.
Slow rotation approximation
R3Ω2
GM 1 Slow rotation condition, Hartle (1967)
Extreme cases (millisecond pulsars, M = 1.4M, R = 10km):
• J1748-2446ad, Ω = 4501 rad s−1 (not seen glitching):
R3Ω2/(GM) ≈ 0.11
• J1824-2452A, Ω = 2057 rad s−1 (seen glitching, Cognard & Backer 2004):
R3Ω2/(GM) ≈ 0.023
Mass upper limit Mmax, relativistic corrections
The calculation of the maximum glitch has been remade in the slow rotationframework in Antonelli, Montoli & Pizzochero (2018).
The dependence of Mmax on relativistic corrections is weak.
0.8 1.0 1.2 1.4 1.6 1.8 2.0M(MSun)
0
2
4
6
8
Maxim
um
glit
ch a
mplit
ude (
1e-4
rad
/s)
EOS: SLy
Newtonian (N)
Rigid vortices (R)
Slack vortices (S)
0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0Maximum glitch amplitude (1e-4 rad/s)
0.00
0.01
0.02
0.03
0.04
0.05
Mm
ax(R
,S)
/ M
max(N
) -
1
Rigid vortices (R)
SLy4
BSk20
BSk21
SLy4
BSk20
BSk21
Slack vortices (S)
Maximum glitch amplitude
Maximal glitch
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0M
axim
al g
litch
(1e-
4 ra
d/s)
Crust Nmax < 3Nmax < 4Nmax > 4
0.60 n0 0.68 n0
10 3 10 2 10 1 100
Nominal lag (rad/s)
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
Max
imal
glit
ch (1
e-4
rad/
s)
0.75 n0
10 3 10 2 10 1 100
Nominal lag (rad/s)
1 n0
10 3 10 2 10 1 100
Nominal lag (rad/s)
Whole star
Pulsars have been chosen with Ngl > 2, Nmax > 1.1, Tobs |Ω∞|> 10−3 rad/s.
Metric of a slowly rotating star
ds2 = −e2Φ(r)dt2 + e2Λ(r)dr2 + r2 [dϑ2 + sin2 ϑ(dϕ− ω(r)dt)2]The radial profiles are still given by the TOV equations + EoS. ω(r) is given byan additional differential equation.
Drag of the local inertial frames
0.55
0.6
0.65
0.7
0.75
0.8
0.85
0.9
0 2 4 6 8 10 12 14
ω‾ (
r)/Ω
r [km]
1.4 MSun SLy41.4 MSun BSk211.4 MSun BSk20
The relevant quantity is ω(r) = Ω− ω(r).
Due to the linearity of the equation giving ω(r), ω(r)/Ω does not depend on Ω.
Moments of inertia
Total moment of inertia:
I =8π3
∫ R
0dr r4e−Φ(r)+Λ(r)
(ρ(r) +
P(r)
c2
)ω(r)
Ω
Moment of inertia of the superfluid component:
Iv =8π3
∫ R
0dr r4e−Φ(r)+Λ(r)xn(r)
(ρ(r) +
P(r)
c2
)mn
m∗n (r)
Moments of inertia
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
5.0
1 1.5 2
SLy
Mom
ent
of
inert
ia (
10
45
g c
m2)
M (MSun)
Itot (N)Itot (GR)In (N)In (GR)
1 1.5 2
BSk21
M (MSun)
1 1.5 2 2.5
BSk20
M (MSun)
• The total moment of inertia varies up to 50%.
• In GR the moment of inertia of the superfluid reservoir exceeds the totalone.