07.01.2014 Views

PHYS08200604018 Shamik Banerjee - Homi Bhabha National ...

PHYS08200604018 Shamik Banerjee - Homi Bhabha National ...

PHYS08200604018 Shamik Banerjee - Homi Bhabha National ...

SHOW MORE
SHOW LESS

Create successful ePaper yourself

Turn your PDF publications into a flip-book with our unique Google optimized e-Paper software.

50 CHAPTER 4. GENERALITIES OF QUARTER BPS DYON PARTITION FUNCTION<br />

respectively, acting on the original variables (ˇρ, ˇσ, ˇv). Again we could hope that a part of this<br />

symmetry is a symmetry of ̂Φ.<br />

We shall illustrate these by several examples in §4.5.<br />

4.4 Black hole entropy<br />

Another set of constraints may be derived by requiring that the formula for the index of quarter<br />

BPS states match the entropy of the black hole carrying the same charges in the limit when<br />

the charges are large. The consequences of this constraint have been analyzed in detail in the<br />

past [7, 8, 12, 15] and reviewed in [24]. Hence our discussion will be limited to a review of the<br />

salient features.<br />

In the approximation where we keep the supergravity part of the action containing only<br />

the two derivative terms, the black hole entropy is given by<br />

π √ Q 2 P 2 − (Q · P ) 2 . (4.4.1)<br />

In all known cases this result is reproduced by the asymptotic behaviour of (5.0.14) for large<br />

charges. Furthermore the leading asymptotic behaviour comes from the residue of the partition<br />

function at the pole at [7, 8, 12, 15, 24]<br />

ˇρˇσ − ˇv 2 + ˇv = 0 , (4.4.2)<br />

up to translations of ˇρ, ˇσ and ˇv by their periods. We shall assume that this result continues<br />

to hold in the general case. Thus ̂Φ(ˇρ, ˇσ, ˇv) must have a zero at (4.4.2). In order to find<br />

the behaviour of ̂Φ near this zero one needs to know the first non-leading correction to the<br />

leading formula (4.4.1) for the black hole entropy. A priori these corrections depend on the<br />

complete set of four derivative terms in the quantum effective action of the theory and are<br />

difficult to calculate. However in all known examples one finds that the entropy calculated<br />

just by including the Gauss-Bonnet term in the effective action reproduces correctly the first<br />

non-leading correction to the statistical entropy. If we assume that this result continues to<br />

hold for a general theory then we can use this to determine the behaviour of ̂Φ near (4.4.2) in<br />

terms of the coefficient of the Gauss-Bonnet term in the effective action.<br />

Since this procedure has been extensively studied in [7,8,12,15] and reviewed in [24], we<br />

shall only quote the result. Typically the Gauss Bonnet term in the Lagrangian has the form<br />

∫<br />

d 4 x √ − det g φ(a, S) { R µνρσ R µνρσ − 4R µν R µν + R 2} , (4.4.3)<br />

where τ = a + iS is the axion-dilaton modulus and the function φ(a, S) has the form<br />

φ(a, S) = − 1 ((k + 2) ln S + ln g(a + iS) + ln g(−a + iS)) + constant . (4.4.4)<br />

64π2

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!