PDA

View Full Version : central charges


Tobias Sander
Nov19-04, 03:38 PM
<jabberwocky><div class="vbmenu_control"><a href="jabberwocky:;" onClick="newWindow=window.open('','usenetCode','toolbar=no, location=no,scrollbars=yes,resizable=yes,status=no ,width=650,height=400'); newWindow.document.write('<HTML><HEAD><TITLE>Usenet ASCII</TITLE></HEAD><BODY topmargin=0 leftmargin=0 BGCOLOR=#F1F1F1><table border=0 width=625><td bgcolor=midnightblue><font color=#F1F1F1>This Usenet message\'s original ASCII form: </font></td></tr><tr><td width=449><br><br><font face=courier><UL><PRE>Some time ago, one Rufus Anton asked in this forum about the seemingly\nmissing central charge of D6 branes:\n\nhttp://groups.google.com/groups?hl=en&lr=&threadm=a1c70df8.0409211643.364872-100000%40posting.google.com&rnum=2&prev=/groups%3Fq%3DRufus%2BAnton%26hl%3Den%26lr%3D%26gro up%3Dsci.physics.strings%26selm%3Da1c70df8.0409211 643.364872-100000%2540posting.google.\ncom%26rnum%3D2\n\nThis fact is known to many string theorists, but it continues to\npuzzle me a little bit. Let me explain -- here is the story:\n\n* Firstly, look at 11D SUGRA\n\nThe 11d SUSY algebra is schematically\n\n{Q,Q} = P_\\mu \\Gamma^\\mu +Z^2_{...} \\Gamma^{...} + Z^5_{...} \\Gamma^{...}\n\nHere P_\\mu is the 11d momentum, Z^2 is the central charge of the\nM2-brane and, Z^5 that of the M5-brane. (I have suppressed any spinor\nindices and the dots ... indicate contraction of the spacetime indices\nwith the appropriate \\Gamma-matrices.)\n\nNow, if you reduce on a circle, you get a scalar and a 10-vector from\nP_\\mu, which you can recognize as the D0-central charge and the\n10-momentum respectively. Similarly, you get a 1-form and a 2-form\nfrom Z_2, being the F1-charge and the D2-charge. And from Z_5 you get\na 4-form and a 5-form, the D4-charge and the NS5 charge.\n\nIn this scheme D(-1), D6, and D8 do not appear to have a charge. For\nD(-1) and D8 this is not surprising to me -- but for D6 it is at first\nsight. Lubos gave the answer to this in his reply to Rufus\' question,\npointing out that the D6 becomes a KK-monopole at strong coupling.\n\n[Moderator\'s note: You should first decide whether you talk about type\nIIA picture, or type IIB. It\'s no good to mix odd numbers like D(-1)\nwith even numbers like D6 and D8. Moreover, the D(-1)charge is not\na conserved operator - it is a feature of a spacetime configuration\n(D-instanton, localized both in space and the Euclidean time)\nthat does not survive in the Hilbert space formalism, I think. LM]\n\n* Secondly, let\'s look at this from the 10d perspective directly. For\nIIA there are 32 supercharges organized in two spinors Q and {\\bar Q}\nwhich must be in the 16 and 16\' of SO(9,1), respectively. The possible\ncentral charges are therefore in the product representations of 16 and\n16\', namely:\n\n(16 \\times 16\') = (16\' \\times 16) = [0] + [2] + [4] (=&gt;\nD0,D2,D4-charges)\n(16 \\times 16)_s = [1] + [5]_+ (=&gt; F1,NS5-charges)\n(16\' \\times 16\')_s = [1] + [5]_- (=&gt; F1,NS5-charges)\n\n(The subscript _s makes clear that the anticommutator we consider has\ndefinite permuational symmetry with respect to its spinor indices,\nwhich explains the absence of tensors like [3].)\n\nAgain, D(-1), D6, D8-charges do not appear. Notice that if you count\nthe degrees of freedom of all the tensors on the RHS, you find 528.\nThis saturates the number of degrees of freedom that we expect for a\ntheory with 32 supercharges, because 32*33/2=528. That is, the algebra\nis complete without the missing D(-1), D6, D8-charges.\n\n* Thirdly, look at the same story in IIB-theory. Here you have two\nspinors of same chirality, let\'s call them Q_A and Q_B, both transform\nin either 16 or 16\'. The product representations are therefore\n\n(16_A \\times 16_B) = (16_B \\times 16_A) = [1] + [3] + [5]_+ (=&gt;\nD1,D3,D5-charges)\n(16_A \\times 16_A)_s = (16_B \\times 16_B)_s = [1] + [5]_+ (=&gt;\nF1,NS5-charges)\n\nNow, the D(-1), D7 and D9-branes do not appear to have charges. This\nis no surprise to me, given the nature of these objects. Again, the\ntotal number of degrees of freedom comes out all right (528).\n\n* Finally my question: The central charge of IIB\'s D5-brane *does*\nappear in the algebra, while that of IIA\'s D6-brane *does not*. But\nthe two are T-dual to one another, are they not? At the very least,\nthis seems to imply a quite non-trivial mapping of the central\ncharges, and perhaps the supercharges, under T-duality. Is this\nmapping known explicitly? Am I right to be confused about this?\n\nThanks,\nTobias\n\n</UL></PRE></font></td></tr></table></BODY><HTML>');"> <IMG SRC=/images/buttons/ip.gif BORDER=0 ALIGN=CENTER ALT="View this Usenet post in original ASCII form">&nbsp;&nbsp;View this Usenet post in original ASCII form </a></div><P></jabberwocky>Some time ago, one Rufus Anton asked in this forum about the seemingly
missing central charge of D6 branes:

http://groups.google.com/groups?hl=en&lr=&threadm=a1c70df8.0409211643.364872-100000%40posting.google.com&rnum=2&prev=/groups%3Fq%3DRufus%2BAnton%26hl%3Den%26lr%3D%26gro up%3Dsci.physics.strings%26selm%3Da1c70df8.0409211 643.364872-100000%2540posting.google.
com%26rnum%3D2

This fact is known to many string theorists, but it continues to
puzzle me a little bit. Let me explain -- here is the story:

* Firstly, look at 11D SUGRA

The 11d SUSY algebra is schematically

{Q,Q} = P_\mu \Gamma^\mu +Z^{2_}{...} \Gamma^{...} + Z^{5_}{...} \Gamma^{...}

Here P_\mu is the 11d momentum, Z^2 is the central charge of the
M2-brane and, Z^5 that of the M5-brane. (I have suppressed any spinor
indices and the dots ... indicate contraction of the spacetime indices
with the appropriate \Gamma-matrices.)

Now, if you reduce on a circle, you get a scalar and a 10-vector from
P_\mu, which you can recognize as the D0-central charge and the
10-momentum respectively. Similarly, you get a 1-form and a 2-form
from Z_2, being the F1-charge and the D2-charge. And from Z_5 you get
a 4-form and a 5-form, the D4-charge and the NS5 charge.

In this scheme D(-1), D6, and D8 do not appear to have a charge. For
D(-1) and D8 this is not surprising to me -- but for D6 it is at first
sight. Lubos gave the answer to this in his reply to Rufus' question,
pointing out that the D6 becomes a KK-monopole at strong coupling.

[Moderator's note: You should first decide whether you talk about type
IIA picture, or type IIB. It's no good to mix odd numbers like D(-1)
with even numbers like D6 and D8. Moreover, the D(-1)charge is not
a conserved operator - it is a feature of a spacetime configuration
(D-instanton, localized both in space and the Euclidean time)
that does not survive in the Hilbert space formalism, I think. LM]

* Secondly, let's look at this from the 10d perspective directly. For
IIA there are 32 supercharges organized in two spinors Q and {\bar Q}
which must be in the 16 and 16' of SO(9,1), respectively. The possible
central charges are therefore in the product representations of 16 and
16', namely:

(16 \times 16') = (16' \times 16) = [0] + [2] + [4] (=>
D0,D2,D4-charges)
(16 \times 16)_s = [1] + [5]_+ (=> F1,NS5-charges)
(16' \times 16')_s = [1] + [5]_- (=> F1,NS5-charges)

(The subscript _s makes clear that the anticommutator we consider has
definite permuational symmetry with respect to its spinor indices,
which explains the absence of tensors like [3].)

Again, D(-1), D6, D8-charges do not appear. Notice that if you count
the degrees of freedom of all the tensors on the RHS, you find 528.
This saturates the number of degrees of freedom that we expect for a
theory with 32 supercharges, because 32*33/2=528. That is, the algebra
is complete without the missing D(-1), D6, D8-charges.

* Thirdly, look at the same story in IIB-theory. Here you have two
spinors of same chirality, let's call them Q_A and Q_B, both transform
in either 16 or 16'. The product representations are therefore

(16_A \times 16_B) = (16_B \times 16_A) = [1] + [3] + [5]_+ (=>
D1,D3,D5-charges)
(16_A \times 16_A)_s = (16_B \times 16_B)_s = [1] + [5]_+ (=>
F1,NS5-charges)

Now, the D(-1), D7 and D9-branes do not appear to have charges. This
is no surprise to me, given the nature of these objects. Again, the
total number of degrees of freedom comes out all right (528).

* Finally my question: The central charge of IIB's D5-brane *does*
appear in the algebra, while that of IIA's D6-brane *does not*. But
the two are T-dual to one another, are they not? At the very least,
this seems to imply a quite non-trivial mapping of the central
charges, and perhaps the supercharges, under T-duality. Is this
mapping known explicitly? Am I right to be confused about this?

Thanks,
Tobias

Lubos Motl
Nov19-04, 04:07 PM
<jabberwocky><div class="vbmenu_control"><a href="jabberwocky:;" onClick="newWindow=window.open('','usenetCode','toolbar=no, location=no,scrollbars=yes,resizable=yes,status=no ,width=650,height=400'); newWindow.document.write('<HTML><HEAD><TITLE>Usenet ASCII</TITLE></HEAD><BODY topmargin=0 leftmargin=0 BGCOLOR=#F1F1F1><table border=0 width=625><td bgcolor=midnightblue><font color=#F1F1F1>This Usenet message\'s original ASCII form: </font></td></tr><tr><td width=449><br><br><font face=courier><UL><PRE>Hey Tobias,\n\nall charges of BPS branes appear in the SUSY algebra as long as the branes\nare "small" perturbations of the background, I think. If you have one of\nyour "unknown" branes stretched in directions {D}, then it will contribute\nto the central charge associated with the remaining directions *{D}. For\nexample, a D6-brane contributes by \\Gamma_{abcdef} to the RHS of the SUSY\nalgebra, which is equivalent to its product with the helicity gamma matrix\ni.e. to \\Gamma_{11}\\Gamma_{abcdef}.\n\nConcretely, a D6-brane extended in time and wrapped in 1,2,3,4,5,6\ndirections contributes like a "spatial" D4-brane wrapped in the 0,7,8,9\ndirections. However, one must discuss every case separately if you want to\nsee the subtleties in M-theory. D8-branes can\'t be really extrapolated\nfrom weakly coupled type IIA to strongly coupled type IIA i.e. M-theory.\nIf you try to do it, the dilaton runs increasingly quickly, and diverges\nalmost immediately unless the total dilaton source is cancelled.\n\nThe way to cancel the dilaton source is to take 8 D8-branes and an\norientifold O8-plane (plus the 8 images of those D8-branes). This has no\ndilaton source, and you can extrapolate it to the strong coupling which\nleads to the Horava-Witten domain wall in M-theory with the E8 gauge\nsymmetry. Be sure that this domain wall preserves 1/2 of the supersymmetry\nin 11 dimensions, namely those components with Gamma_{10}=+1 where x^{10}\nis the transverse dimension fo the domain wall and +1 (or -1) tells you\nthe orientation of the domain wall.\n\nHowever, it does not really make much sense to associate a central charge\nwith the Horava-Witten domain wall because this domain wall is not a small\nperturbation of the infinite spacetime; it changes the asymptotic\nconditions completely.\n\nIn weakly coupled type IIA, on the other hand, there is a central charge\ndescribing D8-branes, and much like D4-branes come with \\Gamma_{abcd},\nD8-branes come with \\Gamma_{abcdefgh}, is there a problem? In the same\nway, weakly coupled type IIA has the required terms for all other 1/2 BPS\nbranes - for D-branes you just insert the appropriate \\Gamma matrix with\n"p" indices. The same thing holds for type IIB, except for D(-1)-branes\nthat are *not* associated with any operator on the Hilbert space because\nthe D(-1)-charge is not a conserved charge (operator) on this Hilbert\nspace but rather a property of configurations in the Euclidean spacetime,\nand you should not insert the D(-1)-branes on the RHS of the SUSY algebra.\n\nThis should clarify the behavior for perturbative type II theories.\nThe M-theory limit is more subtle - and as said above, the D8-branes don\'t\nreally exist as isolated objects. The D6-branes become KK monopoles in\nM-theory, and formally you may imagine that a KK-monopole stretched in the\n1,2,3,4,5,6 directions, being magnetically charged with respect to the x^7\ndirection, contributes with Gamma_{0,7,8,9,10}, much like an M5-brane that\nwould be wrapped in the time direction "twice". I hope that others will\ncorrect me if what I write is wrong.\n\nCheers\nLubos\n_________________________ __________________________________________________ ___\nE-mail: lumo@matfyz.cz fax: +1-617/496-0110 Web: http://lumo.matfyz.cz/\neFax: +1-801/454-1858 work: +1-617/384-9488 home: +1-617/868-4487 (call)\nWebs: http://schwinger.harvard.edu/~motl/ http://motls.blogspot.com/\n^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n\n\n</UL></PRE></font></td></tr></table></BODY><HTML>');"> <IMG SRC=/images/buttons/ip.gif BORDER=0 ALIGN=CENTER ALT="View this Usenet post in original ASCII form">&nbsp;&nbsp;View this Usenet post in original ASCII form </a></div><P></jabberwocky>Hey Tobias,

all charges of BPS branes appear in the SUSY algebra as long as the branes
are "small" perturbations of the background, I think. If you have one of
your "unknown" branes stretched in directions {D}, then it will contribute
to the central charge associated with the remaining directions *{D}. For
example, a D6-brane contributes by \Gamma_{abcdef} to the RHS of the SUSY
algebra, which is equivalent to its product with the helicity \gamma matrix
i.e. to \Gamma_{11}\Gamma_{abcdef}.

Concretely, a D6-brane extended in time and wrapped in 1,2,3,4,5,6
directions contributes like a "spatial" D4-brane wrapped in the 0,7,8,9
directions. However, one must discuss every case separately if you want to
see the subtleties in M-theory. D8-branes can't be really extrapolated
from weakly coupled type IIA to strongly coupled type IIA i.e. M-theory.
If you try to do it, the dilaton runs increasingly quickly, and diverges
almost immediately unless the total dilaton source is cancelled.

The way to cancel the dilaton source is to take 8 D8-branes and an
orientifold O8-plane (plus the 8 images of those D8-branes). This has no
dilaton source, and you can extrapolate it to the strong coupling which
leads to the Horava-Witten domain wall in M-theory with the E8 gauge
symmetry. Be sure that this domain wall preserves 1/2 of the supersymmetry
in 11 dimensions, namely those components with \Gamma_{10}=+1 where x^{10}
is the transverse dimension fo the domain wall and +1 (or -1) tells you
the orientation of the domain wall.

However, it does not really make much sense to associate a central charge
with the Horava-Witten domain wall because this domain wall is not a small
perturbation of the infinite spacetime; it changes the asymptotic
conditions completely.

In weakly coupled type IIA, on the other hand, there is a central charge
describing D8-branes, and much like D4-branes come with \Gamma_{abcd},
D8-branes come with \Gamma_{abcdefgh}, is there a problem? In the same
way, weakly coupled type IIA has the required terms for all other 1/2 BPS
branes - for D-branes you just insert the appropriate \Gamma matrix with
"p" indices. The same thing holds for type IIB, except for D(-1)-branes
that are *not* associated with any operator on the Hilbert space because
the D(-1)-charge is not a conserved charge (operator) on this Hilbert
space but rather a property of configurations in the Euclidean spacetime,
and you should not insert the D(-1)-branes on the RHS of the SUSY algebra.

This should clarify the behavior for perturbative type II theories.
The M-theory limit is more subtle - and as said above, the D8-branes don't
really exist as isolated objects. The D6-branes become KK monopoles in
M-theory, and formally you may imagine that a KK-monopole stretched in the
1,2,3,4,5,6 directions, being magnetically charged with respect to the x^7
direction, contributes with \Gamma_{0,7,8,9,10}, much like an M5-brane that
would be wrapped in the time direction "twice". I hope that others will
correct me if what I write is wrong.

Cheers
Lubos
__{_______________________________________________ _____________________________}
E-mail: lumo@matfyz.cz fax: +1-617/496-0110 Web: http://lumo.matfyz.cz/
eFax: +1-801/454-1858 work: +1-617/384-9488 home: +1-617/868-4487 (call)
Webs: http://schwinger.harvard.edu/~motl/ http://motls.blogspot.com/
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ ^^^^^^^^^^^^^^^^^^^^^^^^^^^^

Jack Tremarco
Nov20-04, 12:26 PM
<jabberwocky><div class="vbmenu_control"><a href="jabberwocky:;" onClick="newWindow=window.open('','usenetCode','toolbar=no, location=no,scrollbars=yes,resizable=yes,status=no ,width=650,height=400'); newWindow.document.write('<HTML><HEAD><TITLE>Usenet ASCII</TITLE></HEAD><BODY topmargin=0 leftmargin=0 BGCOLOR=#F1F1F1><table border=0 width=625><td bgcolor=midnightblue><font color=#F1F1F1>This Usenet message\'s original ASCII form: </font></td></tr><tr><td width=449><br><br><font face=courier><UL><PRE>Lubos Motl wrote:\n\n&gt; Concretely, a D6-brane extended in time and wrapped in 1,2,3,4,5,6\n&gt; directions contributes like a "spatial" D4-brane wrapped in the 0,7,8,9\n&gt; directions. [...]\n\nI had a hard time understanding the two previous posts, but here you\nreally lost me. A D6 brane is extended in time and 6 spatial\ndimensions, say 0123456. The remaining directions are three spatial\ndimensions 789, corresponding to a three dimensional object localized\nin time (an S3-brane?). Even if I somehow miss one direction, can you\nexplain how a D4-brane and an S4-brane are related in the sense of the\npresent discussion?\n\n--Jack\n\n[Moderator\'s note: E.g. a D4-brane is extended in the time direction, but\nthe contribution to the RHS of the SUSY algebra, written covariantly as\n{Q^a, Qbar_b}, is proportional to the gamma matrix that only contains\nthe spatial indices, not the time index. I was trying to explain which\nindices appear in the appropriate gamma matrices on the RHS of the SUSY\nalgebra. I\'ve never used the word "S-brane", and it does not seem as\na correct interpretation of the situation to me. There are no S-branes\nin this game. All the branes, except for the D-instantons that don\'t\nappear in the SUSY algebra, are stretched in time, nevertheless the\nindex 0 sometimes appears in the gamma matrix, and sometimes it does\nnot - it\'s just a matter of Hodge-dualization. A simple result\nof the fact that \\Gamma_{0,1,2,3,4,5,6,7,8,9,(10)} containing all\nthe indices simply acts as +1 or -1 on the supercharges. LM]\n\n</UL></PRE></font></td></tr></table></BODY><HTML>');"> <IMG SRC=/images/buttons/ip.gif BORDER=0 ALIGN=CENTER ALT="View this Usenet post in original ASCII form">&nbsp;&nbsp;View this Usenet post in original ASCII form </a></div><P></jabberwocky>Lubos Motl wrote:

> Concretely, a D6-brane extended in time and wrapped in 1,2,3,4,5,6
> directions contributes like a "spatial" D4-brane wrapped in the 0,7,8,9
> directions. [...]

I had a hard time understanding the two previous posts, but here you
really lost me. A D6 brane is extended in time and 6 spatial
dimensions, say 0123456. The remaining directions are three spatial
dimensions 789, corresponding to a three dimensional object localized
in time (an S3-brane?). Even if I somehow miss one direction, can you
explain how a D4-brane and an S4-brane are related in the sense of the
present discussion?

--Jack

[Moderator's note: E.g. a D4-brane is extended in the time direction, but
the contribution to the RHS of the SUSY algebra, written covariantly as
{Q^a, Qbar_b}, is proportional to the \gamma matrix that only contains
the spatial indices, not the time index. I was trying to explain which
indices appear in the appropriate \gamma matrices on the RHS of the SUSY
algebra. I've never used the word "S-brane", and it does not seem as
a correct interpretation of the situation to me. There are no S-branes
in this game. All the branes, except for the D-instantons that don't
appear in the SUSY algebra, are stretched in time, nevertheless the
index sometimes appears in the \gamma matrix, and sometimes it does
not - it's just a matter of Hodge-dualization. A simple result
of the fact that \Gamma_{0,1,2,3,4,5,6,7,8,9,(10)} containing all
the indices simply acts as +1 or -1 on the supercharges. LM]