Accretion in dwarf novae Nicolas Scepi supervised by Guillaume - - PowerPoint PPT Presentation

Accretion in dwarf novae

Nicolas Scepi

supervised by Guillaume Dubus and Geoffroy Lesur

Dautreppe, 4th of December 2018

1

Dwarf novae

2 White dwarf Solar type star Accretion disk

Dwarf novae are ideal to study accretion :

• emission in the visible, UV
• access to structure of the disk via eclipse mapping
• high variability with time scales going from seconds to

months

3

Cannizzo et al 2010

Days

M a g n i t u d e

Variability in dwarf novae (DNe)

Luminosity coming from the accretion in the disk.

(Shakura & Sunyaev 1973)

Historical framework : Turbulent/viscous accretion

Accretion disk

Angular momentum transport

Companion

WD

Accretion

turbulence

Turbulent transport modeled as a viscous transport (Shakura & Sunyaev 1973)

4

νeff = αcsH

where turbulence is supposedly due to MRI. (Balbus & Hawley 1991)

5

Disk instability model (DIM)

Cannizzo et al 2010 Days Magnitude S-curve from the DIM

· Mexternal

Mass accretion rate (or Temperature)

Eruptive state ⍺ ~ 0.1

(Kotko & Lasota 2012)

Quiescent state ⍺ ~ 0.01

(Cannizzo et al. 2012)

tvis = 1 αΩ ( R H )

2

ttherm = 1 αΩ

6

Disk instability model (DIM)

Cannizzo et al 2010

Days Magnitude S-curve from the DIM Mass accretion rate

· Mexternal

Eruptive state ⍺ ~ 0.1

(Kotko & Lasota 2012)

Quiescent state ⍺ ~ 0.01

(Cannizzo et al. 2012)

tvis = 1 αΩ ( R H )

2

ttherm = 1 αΩ

Can MRI give these values of ⍺ ?

7

Shearing box simulations

Compute ⍺ from the simulations !!

8

Magnetic configuration

Bz Zero Net Flux (ZNF) Net Flux Bz

9

Magnetic configuration

Bz Bz

⍺ does not depend on Bz

(Hawley et al. 1996,Simon et al. 2012)

⍺ depends on Bz

(Hawley et al. 1995)

Zero Net Flux (ZNF) Net Flux

10

1) Zero Net Flux simulations 2) Net Flux simulations 3) Disk-wind model Overview

11

1) Zero Net Flux simulations

2) Net Flux simulations 3) Disk-wind model Overview

12

Do not match observational light curves !

Light curves from Zero Net Flux simulations

Using ⍺~0.1 for eruptive state ⍺~0.01 for quiescent state Using ⍺ from simulations

(Hirose et al. 2014, Scepi et al. 2018a)

Coleman et al. 2016 Coleman et al. 2016

13

Resistive cold branch

Scepi et al. 2018a

When we include resistivity MRI is quenched in the quiescent state (as predicted by Gammie & Menou 1998). Yet, there is observational evidence that DNe in quiescence accrete (Mukai et al 2017).

14

1) Zero Net Flux simulations

2) Net Flux simulations

3) Disk-wind model Overview

15

Net flux simulations

Accretion disk

Transport of angular momentum

Companion

WD

Accretion turbulence

· MRϕ

· M

=

Mass accretion rate due to turbulent transport.

{

16

Outflows

Accretion disk

Transport of angular momentum

Companion

WD

Accretion turbulence

• utflows

· MRϕ

· Mzϕ

{ {

· M

= +

Mass accretion rate due to turbulent transport. Mass accretion rate due to wind-driven transport.

Turbulent VS wind-driven accretion

17

10−2

H/R

102 103 104 105 106 107

β

Viscously driven Wind-driven

C = 6, β < 104 C = 3.5, β < 104 C = 2, β > 104

−0.5 −0.4 −0.3 −0.2 −0.1 0.0 0.1 0.2 0.3 0.4 0.5

log10(

˙ MRφ ˙ MZφ)

Eruptive state dominated by viscous accretion Quiescent state dominated by the wind- driven accretion

Scepi et al. 2018b

A new framework

18

Accretion disk

Transport of angular momentum Companion

WD

Accretion turbulence

• utflows

Disk with a wind will not behave as an ⍺-disk. Need to review observational constraints with a disk-wind model.

Accretion disk

Angular momentum transport

Companion

WD

Accretion

turbulence

19

1) Zero Net Flux simulations 2) Net Flux simulations

3) Disk-wind model

Overview

20

A new disk-wind instability model

Accretion disk

Transport of angular momentum

Companion

WD

Accretion turbulence

• utflows

We used prescriptions on from our simulations to construct a new DIM. We used a fixed magnetic field configuration.

α(β), q(β)

21

B dipolar

Stable case

· Mexternal = 3 × 1017 g s−1, Rout = 2 × 1010 cm

Scepi et al. 2018c in prep

22

Scepi et al. 2019 in prep

· Mexternal = 1 × 1016 g s−1, Rout = 2 × 1010 cm

B dipolar

Unstable case

Scepi et al. 2018c in prep

23

Observations vs Model

Scepi et al. 2018c in prep

Cannizzo et al 2010

Days

Magnitude

For a dipolar moment of ~1030 G cm3, light curves are very similar to that of DNe! However, we used a fixed magnetic

• field. We need to compute the

evolution of the magnetic field.

Conclusions

24

• Turbulent MRI transport alone cannot explain the

behavior of DNe

• Net Flux simulations show that outflows transport angular

momentum very efficiently in the quiescent state

• Taking into account turbulent and wind-driven transport, we

can reproduce light-curves of DNe

Thank you for your attention

25