|Views: 155|Likes: 0

Published by Markus Johan Kiril Klyver

See More

See less

L87

The Astrophysical Journal

, 591:L87–L89, 2003 July 10

2003. The American Astronomical Society. All rights reserved. Printed in U.S.A.

PROBING PLANCK-SCALE PHYSICS WITH EXTRAGALACTIC SOURCES?

Y. Jack Ng, W. A. Christiansen, and H. van Dam

Department of Physics and Astronomy, University of North Carolina, CB 3255, Phillips Hall, Chapel Hill, NC 27599;yjng@physics.unc.edu, wayne@astro.unc.edu

Received 2003 March 21; accepted 2003 May 27; published 2003 June 16

ABSTRACTAt Planck scale, spacetime is “foamy” because of quantum ﬂuctuations predicted by quantum gravity. Herewe consider the possibility of using spacetime foam-induced phase incoherence of light from distant galaxiesand gamma-ray bursters to probe Planck-scale physics. In particular, we examine the cumulative effects of spacetime ﬂuctuations over a huge distance. Our analysis shows that they are far below what is required in thisapproach to shed light on the foaminess of spacetime.

Subject headings:

distance scale—galaxies: general— gamma rays: bursts —gravitation—radiation mechanisms: general— techniques: interferometricIt is generally believed that quantum gravity, the synthesis of quantum mechanics and general relativity, predicts that space-time becomes “foamy” or “fuzzy” at the Planck scale given bythe Planck time , Planck length

5 1/2

44

t

p

(

G

/

c

)

∼

10 s

l

p

P P

, and Planck energy . The

33 28

ct

∼

10 cm

E

p

/

t

∼

10 eV

P P P

fuzziness of spacetime leads to uncertainties in distance (

l

) mea-surementswhoseabsolutevalueisgivenby (sim-

1

a

d

l

l

(

l

/

l

)

P P

ilar uncertainties for time measurements) and uncertainties inenergy (

E

) measurements given by (again

d

E

is

a

d

E

E

(

E

/

E

)

P

an absolute value). There are also similar uncertainties for mo-mentum measurements (Ng & van Dam 1994). The parameterspeciﬁes different quantum gravity models. The standard

a

∼

1choice (Misner, Thorne, & Wheeler 1973, p. 1190) of

a

is; the choice of appears (Ng & van Dam 2000; Ng

2

a

p

1

a

p

3

2001) to be consistent with the holographic principle (’t Hooft1993; Susskind 1995) and black hole physics; cor-

1

a

p

2

responds to the random walk model found in the literature(Amelino-Camelia 1994).The ultra–high-energy cosmic-ray threshold anomalies (see,e.g., Amelino-Camelia & Piran 2001; Ng et al. 2001; Amelino-Camelia,Ng,&vanDam2002;Aloisioetal.2003andreferencestherein) may have given us some tantalizing hints of Planckianﬂuctuations. But so far we lack direct experimental evidence.Recently, Lieu & Hillman (2002, 2003) and then Ragazzoni,Turatto, &Gaessler(2003)proposedatechniquethathashithertobeen overlooked to directly test Planck-scale ﬂuctuations. Theyarguedthattheseﬂuctuationscancumulativelyleadtoacompleteloss of phase coherence for radiation that has propagated a suf-ﬁciently large distance. As a result of this inferred phase scram-bling attributed to Planck-scale uncertainties (spacetime foam),these authors concluded that distant compact radiation sourcesshould not produce the normal interference patterns (e.g., Airyrings) that are often observed. In this Letter, we examine theirvery interesting idea and extend their proposal to include theuse of gamma-ray burst interferometry.Since the cumulative effect of quantum foam on the phasecoherence of light from extragalactic sources ﬁgures cruciallyin the method, we begin by establishing how large the cu-mulative effects are, especially for the three quantum gravitymodels mentioned above.Consider a distance

L

(the reader can anticipatethistodenotethedistancebetweentheextragalacticsourceandthetelescope).Let us divide it into equal parts, each of which has length

L

/

ll

(the reader can anticipate this to denote the wavelength of the observed light from the distant source). In principle, anylength suitably larger than

l

P

can be used as

l

. But, as shownbelow, the wavelength of the observed light is the naturalchoice. We already know that the absolute value of the un-certainty in distance

L

is given by .

1

a

d

L

∼

l

(

L

/

l

)

P P

But let us calculate it again by starting with

dl

for each partand then adding up the contributions to

d

L

from the parts

L

/

l

in

L

. In doing so, we will ﬁnd out how large the cumulativeeffects from the various

dl

are. To gain insight into the process,let us ﬁrst consider the random walk model of quantum

1

a

p

2

gravity. In this model, the typical quantum ﬂuctuation

dl

foreach segment of length

l

is of the order of and takes

1/2

l

(

l

/

l

)

P P

a

sign with equal probability. (In the terminology used inAmelino-Camelia et al. 2002, this is referred to as a “nonsys-tematic” effect of quantum gravity.) To simplify this part of theargument, let us assume that

dl

takes on only two values, viz.,(instead of, say, a Gaussian distribution about zero,

1/2

l

(

l

/

l

)

P P

which is more likely). If the ﬂuctuations from the different seg-ments were all of the same sign (i.e., coherent), then togetherthey would contribute to

d

L

. But both cases,

1/2

l

(

l

/

l

) (

L

/

l

)

P P

yielding a linear

L

-dependence for

d

L

, are ruled out for inco-herent quantum gravity ﬂuctuations. Each has a probability of for . [We note that for the

L

/

l

30

1/2

K

1 (

L

/

l

)

k

1 (

L

/

l

)

∼

10example involving the active galaxy PKS 1413

135 consideredbelow.] Clearly, we have here a one-dimensional random walk involving steps of equal size (

dl

), each step moving right

L

/

l

or left (corresponding to the plus or minus sign) with equalprobability. The result is well known: the cumulative ﬂuctuationis given by , which is

l

P

(

L

/

l

P

)

1/2

as expected

1/2

d

L

∼

dl

#

(

L

/

l

)for consistency.Alternatively, we can derive this cumulative factor

1/2

(

L

/

l

)for the random walk model of quantum gravity by simply usingtheexpressionsfor

dl

and

d

L

themselves.Thecumulativefactor

C

a

deﬁned by

d

L

p

C

dl

(1)

a

is given, for the random walk ( ) case, by

1

a

p

2

1/2

d

L LC

p p

, (2)

a

p

1/2

( )

dl l

since and . The result is as ex-

1/2 1/2

d

L

∼

l

(

L

/

l

)

dl

∼

l

(

l

/

l

)

P P P P

pected.

L88 NG, CHRISTIANSEN, & VAN DAM Vol. 591But while we have some intuitive understanding of the cu-mulative factor for the random walk model, viz., why

C

a

p

1/2

it scales as the square root of the number of “steps,” we

L

/

l

have much less intuition for the other cases. Nevertheless,sincewe know

d

L

and

dl

, we can use equation (1) to ﬁnd the cu-mulative factors, with the results

1

a

LC

p

, (3)

a

( )

l

in particular,

1/3 0

L LC

p

,

C

p p

1, (4)

a

p

2/3

a

p

1

( ) ( )

l l

for the holographic case and the “standard” case,

2

a

p

a

p

1

3

respectively. Note that is

independent

of

L

. Strange

C

p

1

a

p

1

as it may seem, the result is not unreasonable if we recall, forthis standard model, , independent of

l

. The crucial point

d

l

∼

l

P

to remember is that for the three quantum gravity models underconsideration,

none

of the cumulative factors is linear in. (In fact, according to eq. [3], the cumulative effects are(

L

/

l

)linear in only for the physically unacceptable case of

L

/

l

, for which .) To obtain the correct cumulative fac-

a

p

0

d

l

∼

l

tor (given by eq. [3]) from what we may inadvertently think it is, viz., (independent of

a

), we have to put in the(

L

/

l

)correction factor .

a

(

l

/

L

)With the correct cumulative factors for the various quantumgravity models at hand, we can now examine the prospect of probing Planck-scalephysicsbyobservingverydistantsources.Consider the phase behavior of radiation with wavelength

l

received from a celestial source located at a distance

L

away.Fundamentally, the wavelength deﬁnes the minimum lengthscale over which physical quantities such as phase and groupvelocities (and hence dispersion relations)canbedeﬁned.Thus,theuncertainty in

l

introducedbyspacetimefoamisthestartingpoint for our analysis. A wave will travel a distance equal toits own wavelength

l

in a time , where is the group

t

p

l

/

v v

g g

velocity ofpropagation, andthephaseofthewaveconsequentlychanges by an amount

v

t

v

p p

f

p

2

p

p

2

p

(5)

l

v

g

(i.e., if ), where is the phase velocity of the

v

p

v

,

f

p

2

p

v

p g p

light wave. Quantum gravity ﬂuctuations, however, introducerandom uncertainties into this phase, which is simply

v

p

df

p

2

pd

. (6)

( )

v

g

We now argue that, because of quantum ﬂuctuationsofenergymomentum (Ng & van Dam 1994), and

a

d

E

∼

E

(

E

/

E

)

d

p

∼

P

, the standard radiation dispersion relation

a

2

p

(

pc

/

E

)

E

P

should be changed to . Re-

2 2 2 2 2 2

a

c p

p

0

E

c p

∼

E

(

E

/

E

)

P

calling that and , we obtain

v

p

E

/

p

v

p

dE

/

dp

p g

a a

v

E l

p

P

d

∼

p

, (7)

( )

( ) ( )

v

E

l

P

g

where we have used . We emphasize that this may

E

/

E

p

l

/

l

P P

be either an incremental advance or a retardation in the phase.In traveling over the macroscopically large distance,

L

, fromsource to observer, an electromagnetic wave is continuallysub- jected to random incoherent spacetime ﬂuctuations. Therefore,by our previous argument, the cumulative statistical phase dis-persion is , that is,

Df

p

C

df

aa

1

a a

1

a

l L l L

P P

Df

p

2

p

a

p

2

p

a

, (8)

( ) ( )

l l l

where . This, we believe, is our fundamentaldisagreement

a

∼

1with the Lieu & Hillman paper, in which they assume that themicroscale ﬂuctuations induced by quantum gravity into thephase of electromagnetic waves are coherently magniﬁed bythe factor (see their eq. [11]) rather than .

1

a

L

/

l

(

L

/

l

)In stellar interferometry, following Lieu & Hillman’s (2002,2003) reasoning, for light waves from an astronomical sourceincident upon a two-element interferometer to subsequentlyform interference fringes, it is necessary that . But

Df

2

p

the analysis of the principles of interferometry of distant

in-coherent

astronomical “point” sources can be tricky. The localspatial coherence across an interferometer’s aperture for pho-tons fromadistantpoint source(i.e., planewaves)isareﬂectionof the fact that all photons have the same resultant phase dif-ferences across the interferometer. However, this local coher-ence can be lost if there is an intervening medium such as aturbulent plasma or spacetime foam capable of introducingsmall changes into the “effective” phases of the photon streamfalling on the interferometer. Such spacetime foam-inducedphase differences are themselves incoherent and thereforemustbe treated with the correct cumulative factors

C

a

appropriatefor the quantum gravity model under consideration.Fluctuations due to quantum gravity are very minuscule, sothey can be detected only if there is a huge cumulative effectfrom “summing” up the individual ﬂuctuations. But since thecumulative factor for the standard model of quantum gravity(for which ) is 1, i.e., there is no cumulative effect,

a

p

1obviously the proposed approach (of applying spacetime ﬂuc-tuations on the phase coherence of light from extragalacticsources to probe the graininess of spacetime) cannot be usedto rule out (or conﬁrm) the model. This is the ﬁrst result

a

p

1of this Letter.To rule out models with , the strategy is to to look for

a

!

1interference fringes for which the phase coherence oflightfromthe distant sources should have been lost (i.e., ) for

Df

2

p

that value of

a

according to theoretical calculations. Considerthe example cited by Lieu & Hillman (2003), involving theclearly visible Airy rings in an observation of the active galaxyPKS 1413

135 ( Gpc) by the

Hubble Space Tele- L

p

1.216

scope

(

HST

) at wavelength (Perlman et al. 2002).

l

p

1.6

m

mFor this example, equation (8) yields for the

Df

∼

10

#

2

p

a

random walk model and for the ho-

1

9

a

p

Df

∼

10

#

2

p

a

2

lography model. Since we expect , the observation

2

a

p

a

∼

1

3

of Airy rings in this case would seem to marginally rule outthe random walk model. (But of course proponents of the ran-dom walk model can equally claim that their favorite modelis still marginally acceptable.) On the other hand, the holog-raphy model is obviously not ruled out. This ﬁnding contradictsthe conclusion reached recently by Lieu & Hillman (2003),who argued that the

HST

detection of Airy rings from PKS1413

135 has ruled out a majority of modern models of quan-tum gravity, including the standard model. (Earlier,Lieu

a

p

1