
2
and then taking the D→4 limit of exact solutions
to Einstein-Gauss-Bonnet gravity. This interesting ap-
proach allowed for non-vanishing contributions from the
GB action term in D= 4. In doing so, a number of 4-
dimensional metrics can be obtained (for spherical black
holes [8–12], cosmological solutions [8, 13, 14], star-like
solutions [15, 16], radiating solutions [17], collapsing so-
lutions [18], etc.) carrying imprints of higher curvature
corrections inherited from their D > 4 counterparts.
Unfortunately the existence of limiting solutions does
not actually imply the existence of a 4D theory, and a
number of objections in this vein quickly appeared [19–
21]. However, the conclusion that there ultimately was no
four-dimensional Gauss-Bonnet theory of gravity proved
to be premature when it was shown that a D→4 limit
of the action (2) could be taken [22, 23], generalizing
a previous procedure for taking the D→2 limit of GR
[24]. It is also possible to employ a Kaluza-Klein-like pro-
cedure [25], compactifying D-dimensional Gauss-Bonnet
gravity on a (D−4)-dimensional maximally symmetric
space and then re-scaling the coupling constant according
to equation (2). The resultant 4D scalar-tensor theory is
a special case of Horndeski theory [26], which surprisingly
has spherical black hole solutions whose metric functions
match those from the na¨ıve D→4 limiting solutions de-
rived by Glavan & Lin [8]. Such solutions can be obtained
without ever referencing a higher dimensional spacetime
[22].
This theory, referred to as 4D Einstein Gauss-Bonnet
gravity, provides an interesting phenomenological com-
petitor to GR [27]. Conformally transforming the metric
gµν →e−2ϕgµν in (1) and subtracting it from the original
GB action (1) yields the 4DEGB action after trivial field
redefinitions: [22]
SG
4=αZd4x√−ghϕG+ 4Gµν ∇µϕ∇νϕ−4(∇ϕ)2□ϕ
+ 2(∇ϕ)4i(3)
using (2), where ϕis an additional scalar field. No further
assumptions about particular solutions to higher dimen-
sional theories or background spacetimes are required.
Adding to this the Einstein-Hilbert action with a cos-
mological term,
S=Zd4x√−g[R−2Λ] + SG
4(4)
the equations of motion follow from the standard varia-
tional principle, with that for the scalar being given by
Eϕ=−G + 8Gµν ∇ν∇µϕ+ 8Rµν ∇µϕ∇νϕ−8(□ϕ)2
+ 8(∇ϕ)2□ϕ+ 16∇aϕ∇νϕ∇ν∇µϕ+ 8∇ν∇µϕ∇ν∇µϕ
= 0
(5)
and the variation with respect to the metric yields
Eµν = Λgµν +Gµν +α"ϕHµν −2R[(∇µϕ) (∇νϕ) + ∇ν∇µϕ]
+ 8Rσ
(µ∇ν)∇σϕ+ 8Rσ
(µ∇ν)ϕ(∇σϕ)−2Gµν (∇ϕ)2+ 2□ϕ
−4 [(∇µϕ) (∇νϕ) + ∇ν∇µϕ]□ϕ+ 8 ∇(µϕ∇ν)∇σϕ∇σϕ
−gµν (∇ϕ)2−4 (∇µϕ) (∇νϕ)(∇ϕ)2+ 4 (∇σ∇νϕ) (∇σ∇µϕ)
−4gµν Rσρh∇σ∇ρϕ+ (∇σϕ) (∇ρϕ)i−2gµν (∇σ∇ρϕ) (∇σ∇ρϕ)
−4gµν (∇σϕ) (∇ρϕ) (∇σ∇ρϕ)+4Rµνσρ [(∇σϕ) (∇ρϕ) + ∇ρ∇σϕ]
+ 2gµν (□ϕ)2#= 0
(6)
where His the Gauss-Bonnet tensor:
Hµν = 2hRRµν −2Rµανβ Rαβ +RµαβσRαβσ
ν−2RµαRα
ν
−1
4gµν RαβρσRαβρσ −4Rαβ Rαβ +R2i.
(7)
These field equations satisfy the following relationship
0 = gµν Eµν +α
2Eϕ= 4Λ −R−α
2G(8)
which can act as a useful consistency check to see whether
prior solutions generated via the Glavin/Lin method are
even possible solutions to the theory. For example, using
(8) it is easy to verify that the rotating metrics generated
from a Newman-Janis algorithm [11, 28] are not solutions
to the field equations of the 4DEGB theory.
Despite much exploration of the theory [29], there has
been relatively little work investigating rotating black
hole solutions. Attempts have been made using a na¨ıve
rescaling of the 4DEGB coupling constant [30], or by
implementing the Newman-Janis approach [11, 28, 31],
neither of which produce valid solutions to this theory in
general (the latter case not satisfying the positive energy
condition). A notable exception is a recently obtained
class of asymptotically flat slowly rotating black hole so-
lutions [16] that were obtained for the 4DEGB theory.
However, a detailed study of the geodesics of particles
surrounding such black holes (particularly in de Sitter
and anti-de Sitter spacetimes) was not done.
In this paper we address this issue, obtaining slowly
rotating black hole solutions to the field equations of the
4DEGB theory in asymptotically flat/(A)dS space. We
then analyze their physical properties (innermost stable
circular orbits, photon rings, black hole shadow, etc.)
to see how they differ from standard results in Einstein
gravity. We find for Λ ≤0 that 4DEGB gravity enforces
a minimum mass on the black hole solutions, whereas
for Λ >0 both a minimum and maximum mass occur,
whose horizon structure is directly analogous to that of
Reissner-Nordstr¨om de Sitter spacetime. Besides this,
the results are similar in form to general relativity. The
incredible empirical success of GR necessitates this simi-