Professional Documents
Culture Documents
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
Jaykov Foukzon 1*
jaykovfoukzon@list.ru
Alexander Potapov 2
potapov@cplire.ru
Elena Menkova 3
e_menkova@mail.ru
All-Russian Research Institute for Optical and Physical Measurements, Moscow, Russia
Moscow, 119361
ABSTRACT
This paper dealing with extension of the Einstein field equations using apparatus of
contemporary generalization of the classical Lorentzian geometry named in literature
Colombeau distributional geometry, see for example [1]-[2], [5]-[7] and [14]-[15]. The
regularization of singularities presented in some solutions of the Einstein equations is an
important part of this approach. Any singularities present in some solutions of the Einstein
equations recognized only in the sense of Colombeau generalized functions [1]-[2] and not
classically. In this paper essentially new class Colombeau solutions to Einstein field
equations is obtained. We leave the neighborhood of the singularity at the origin and turn to
the singularity at the horizon. Using nonlinear distributional geometry and Colombeau
generalized functions it seems possible to show that the horizon singularity is not only a
coordinate singularity without leaving Schwarzschild coordinates. However the Tolman
formula for the total energy E T of a static and asymptotically flat spacetime, gives ET m,
as it should be. The vacuum energy density of free scalar quantum field with a
distributional background spacetime also is considered. It has been widely believed that,
except in very extreme situations, the influence of gravity on quantum fields should amount
to just small, sub-dominant contributions. Here we argue that this belief is false by showing
that there exist well-behaved spacetime evolutions where the vacuum energy density of free
quantum fields is forced, by the very same background distributional spacetime such
distributional BHs, to become dominant over any classical energy density component. This
semiclassical gravity effect finds its roots in the singular behavior of quantum fields on
curved spacetimes. In particular we obtain that the vacuum fluctuations
behavior on BHs horizon
r :
2 r ~ r r .
2 has a singular
27
28
29
30
31
32
33
34
35
36
37
38
39
40
Remark 1.1. Note that if some components of the Riemann curvature tensor
41
1. INTRODUCTION
1.1.
The breakdown of canonical formalism of Riemann geometry for the
singular solutions of the Einstein field equations
Einstein field equations was originally derived by Einstein in 1915 in respect with canonical
formalism of Riemann geometry, i.e. by using the classical sufficiently smooth metric tensor,
smooth Riemann curvature tensor, smooth Ricci tensor, smooth scalar curvature, etc.
However, singular solutions of the Einstein field equations with singular metric tensor and
singular Riemann curvature tensor have soon been found. These singular solutions are
formally accepted beyond rigorous canonical formalism of Riemannian geometry.
i
x
Rklm
43
44
remark 1.2.
45
42
46
47
x 0 by formulae
ikl x kl x x i x 0i , 1
kl x C
48
49
and (ii)
50
51
52
x x 0 A k ,
1. 2
53
54
1. 1
0
0
where x x xi xi
i 0
55
contour
56
A k x x 0 A k x x 0 ikl x A k dx l ,
57
58
where
1. 3
Ai
kli x Ak .
l
x
59
1.4
60
61
62
63
64
65
66
67
68
69
70
71
72
73
Now applying Stokes' theorem (see [4], Eq. (6.19)) to the integral (1.3) and considering that
74
f lm , we get
75
A k x x 0 ikl x A k dx l
1
2
ikm x A i x x 0
x l
ikm x A i x x 0
x l
x x 0
ikl x A i x x 0
df lm
x m
ikl x A i x x 0
x m
ikm x x 0 A i
x l
ikm x A i
x x 0
x l
ikl x A i
x x 0
i
x
A
x x
i
kl
x m
x m
0
x x 0
A i x x x 0
76
77
78
ikm x A i
x l
x x 0
f lm
1. 5
f lm
ikl x A i
x m
i
2 ikm x
0 2 kl x
x
x
x
i
x m x 0m
x l x 0l
f lm
.
2
Substituting the values of the derivatives (1.4) into Eq. (1.5), we get finally:
79
i A i x x x 0 f lm
A k Rklm
,
2
80
81
82
where
1. 6
i
Rklm , is a tensor of the fourth rank
83
i
ikm x
ikl x
i
Rklm Rklm 2
.
x l x 0l x m x 0m
84
85
1. 7
i
i
Rklm
is the classical Riemann curvature tensor. That Rklm is a tensor is clear from the
86
Here
87
88
89
at one and the same point. Note that a similar results were obtained by many authors [5]-[17]
by using Colombeau nonlinear generalized functions [1]-[2].
90
91
92
i
Rklm is called the generalized curvature tensor or the generalized
93
94
Rkm is defined as
95
i
Rkm Rkim .
96
97
98
1. 8
R is defined as
99
R g km Rkm .
100
101
102
1. 9
G km is defined as
103
G km Rkm 1 g km R.
2
104
105
1. 10
106
Remark 1.3. (I) Note that the Schwarzschild spacetime is well defined only for r 2m . The
107
108
109
diffeomorfic to a product
horizon [33], [34].
r 2m
in
R 3 R is the submanifold r 2m of R 3 R ,
110
( x0 , r , , ), with m 0,
111
112
113
coefficient
114
115
116
coordinate. Hence the metric cannot be said to be either spherically symmetric or static for
r 2m [33].
117
118
119
120
1 lm
l
kj {} g {} [( g mk , j {} g mj , k {} g kj , m {}]
2
121
1.11
122
123
124
is not available in classical sense and that is well known many years from mathematical
literature, see for example [22] section 6 and Remark 1.1 - Remark 1.2 above.
125
126
127
128
In addition:
129
130
131
132
133
134
135
136
(V) Note that: (i) In physical literature (see for example [4], [33], [35],) it was wrongly
assumed that a properties (i)-(iii) is enough to convince us that r 2m represent an non
honest physical singularity but only coordinate singularity.
137
138
139
140
141
lim Rr 0, Rr
Rr
r2m
r2m
lim Rr 0
r0
r0
............................
lim R rR r
rR r
r2m
r2m
rR r
r0
142
143
144
145
2
48m 2
lim 48m
r6
r6
r2m
2
lim R rR r lim 48m
.
r6
r0
r0
,
r2m
mr 2m
m
1
r r 2 m rlim
0, 11
,
3
2
m
r
r r 2m
m
0
r r 2 m rlim
01
,
2 m r r 2m
1
1
122 r
lim 21 m 1 , 22
lim r 2m 0,
r 2 m
r 2 m
r 2 m r
r 2 m
1
1
133
lim 21 m 1 , 33
lim r 2m sin 2 0,
r 2 m
r 2 m
r 2 m r
r 2 m
mr 2m
m
1
1
r r 0 lim
r r 0 lim
00
11
,
3
r 0
r
0
r
r r 2m
....................................................................
cos
3
332 sin cos , 23
.
sin
1
r
00
146
r 2 m
lim
r 2 m
147
148
149
Thus obviously by consideration above (see Remark 1.1 - Remark 1.2) this extension given
by Eq. (1.15) has no any sense in respect of the canonical Riemannian geometry.
150
(VII) From consideration above (see Remark 1.1 - Remark 1.2) obviously follows that the
151
152
153
154
scalars such as R, R R , . . . , R
respect of the canonical Levi-Civit connection (1.11) and therefore cannot be said to be
either honest physical singularity or only coordinate singularity in respect of the canonical
Riemannian geometry.
155
156
157
158
159
Remark 1.4. Note that in physical literature the spacetime singularity usually is defined as
location where the quantities that are used to measure the gravitational field become infinite
in a way that does not depend on the coordinate system. These quantities are the classical
scalar invariant curvatures of singular spacetime, which includes a measure of the density of
matter.
160
161
Remark 1.5. In general relativity, many investigations have been derived with regard to
singular exact vacuum solutions of the Einstein equation and the singularity structure of
162
T ( x) 0, where
T x represent the energy-momentum densities of the gravity source. This for example is
163
164
the case for the well-known Schwarzschild solution, which is given by, in the Schwarzschild
165
coordinates
( x0 , r , , ),
166
167
rs
,
r
1.12
168
169
where
170
171
172
gravitational constant, mass of the source, and the light velocity in vacuum Minkowski
space-time, respectively. The metric (1.12) describes the gravitational field produced by a
point-like particle located at r 0 .
173
174
Remark 1.6. Note that when we say, on the basis of the canonical expression of the
curvature square
175
R (r )R (r )
176
12rs2
r6
1.13
177
178
179
180
formally obtained from the metric (1.12), that r 0 is a singularity of the Schwarzschild
space-time, the source is considered to be point-like and this metric is regarded as
meaningful everywhere in space-time.
181
182
Remark 1.7. From the metric (1.12), the calculation of the canonical Einstein tensor
proceeds in a straightforward manner gives for r 0
183
184
Gtt r Grr r
hr 1 hr
hr hr
0 , G r G r
2 0,
2
r
r
2
r
1.14
185
186
where
187
G tt 0 lim G tt r 0, G rr 0 lim G rr r 0,
r0
r0
G 0 lim G r 0, G 0 lim G r 0.
r0
188
189
1. 15
r0
190
However as pointed out above the canonical expression of the Einstein tensor in a
191
192
G t 0 lim G t r , G r 0 lim G r r ,
r0
G 0
193
194
195
196
197
lim
r0
r0
G r
, G 0
lim
G r
1. 16
.
r0
and therefore the boundary conditions (1.15) is completely wrong. But other hand as pointed
out by many authors [5]-[17] that the canonical representation of the Einstein tensor, valid
only in a weak (distributional) sense, i.e. [12]:
198
199
200
G ab x 8m a0 0b 3 x
1.17
201
202
203
Einstein tensor is breakdown in rigorous mathematical sense for the Schwarzschild solution
at origin r 0 .
204
205
206
1.2.
207
208
209
210
211
General relativity as a physical theory is governed by particular physical equations; the focus
of interest is the breakdown of physics which need not coincide with the breakdown of
geometry. It has been suggested to describe singularity at the origin as internal point of the
Schwarzschild spacetime, where the Einstein field equations are satisfied in a weak
(distributional) sense [5]-[22].
212
213
214
215
216
217
218
219
The two singular functions we will work with throughout this paper (namely the singular
1
r and
1
1
, rs 0. Since r L1loc (R 3 ),
r rs
G R3 [22]:
220
221
222
223
224
1 1
r
r
1
r
1
r
, 13 r , 0, 1.
1. 18
Inserting (1.18) into (1.12) we obtain a generalized Colombeau object modeling the singular
Schwarzschild spacetime [22]:
225
226
h
r
r
, 0,1.
227
228
229
230
231
232
2
2
2
2
2
ds 2 h rdt 2 h 1
rdr r d sin d
1.19
233
234
235
236
237
238
R r, R
2
2
239
R r, R
0
0
h r 1 h r
r
8m 2 ,
2
r
r
r
1 h r h r
r
4m 2 .
2 2
r
r
3
3
1
1
1.20
240
R r
241
Hence, the distributional Ricci tensor and the distributional curvature scalar
242
-type, i.e. R r m r .
r2
243
Remark 1.9. Note that the formulae (1.20) should be contrasted with what is the expected
244
result
245
obviously given in spherical coordinates and therefore strictly speaking this is not correct,
246
247
248
concentrated at the origin requires a basis regular at the origin. Transforming the formulae
for ( Rij ) into Cartesian coordinates associated with the spherical ones, i.e.,
249
{r , ,} {xi } , we obtain, e.g., for the Einstein tensor the expected result
Gba x 8m 0a b0 3 x given by Eq. (1.17), see [22].
250
are of
, are not globally defined. Representing distributions
,
r
251
252
253
254
255
256
h r
257
1 s r , 0,1, r rs .
r
1.21
258
259
Here
260
261
262
263
264
t
h r 1 h r
r
Gt r , Gr r ,
2
r
r
r
rs
rs 2
2
r
r
1.22
h r h r
2
G r , G r ,
2 r
r r r d r r r .
s
s
s
2
2
r2
r dr
1.23
265
266
267
and
268
269
270
271
which is exactly the result obtained in Ref. [9] using smoothed versions of the Heaviside
272
function
273
274
275
Gba x 8m 0a b0 3 x ,
276
1.24
277
278
279
280
281
282
283
284
1
D(R 3 ) which can be embedded into G R3 [22]:
vp
r
r
s
1
L1loc (R 3 ). A canonical regularization is the principal value
r rs
285
286
287
1 1
1 vp 1
vp
GR3 .
vp
r rs
r rs
r rs r rs
1.25
288
289
290
Inserting now (1.25) into (1.12) we obtain a generalized Colombeau object modeling the
singular Schwarzschild spacetime [22]:
291
hr (dt ) h r (dr )
ds2
292
r 2 (d ) 2 sin 2 (d ) 2 ,
1
r
, 0,1.
hr 1 s , h1 r 1 rs
r
r rs
2
1.26
293
295
r rs ,
because h(r ) is zero at the horizon. However, this does not come as a surprise. Both h(r )
296
and
294
298
h 1 (r ) are positive outside of the black hole and negative in the interior. As a
1
consequence any smooth regularization of h(r ) (or h ) must pass through zero
somewhere and, additionally, this zero must converge to r rs as the regularization
299
300
301
302
Remark 1.11. Note that due to the degeneracy of Colombeau object (1.26), even the
distributional Levi-Civit connection obviously is not available by using the smooth
regularization via horizon [22].
297
303
304
305
306
307
308
309
310
311
In this paper we leave the neighborhood of the singularity at the origin and turn to the
singularity at the horizon. The question we are aiming at is the following: using distributional
geometry (thus without leaving Schwarzschild coordinates), is it possible to show that the
horizon singularity of the Schwarzschild metric is not merely a coordinate singularity. In order
to investigate this issue we calculate the distributional curvature at the horizon in
Schwarzschild coordinates.
312
313
314
315
316
317
318
319
320
321
322
323
324
This of course amounts to the question of whether one can reasonably ascribe distributional
curvature quantities to the Schwarzschild singularity at the horizon.
325
326
The emphasis of the present work lies on mathematical rigor. We derive the physically
expected result for the distributional energy momentum tensor of the Schwarzschild
T00 8m (3) ( x ) , in a conceptually satisfactory way. Additionally, we set up
327
geometry, i.e.,
328
329
330
331
a unified language to comment on the respective merits of some of the approaches taken so
far. In particular, we discuss questions of differentiable structure as well as smoothness and
degeneracy problems of the regularized metrics, and present possible refinements and
workarounds. These aims are accomplished using the framework of nonlinear
332
333
334
1
h(r ) is smooth, h 1 (r ) is not even L loc (note that the origin is now always excluded from
335
336
337
338
339
340
( h
(r ) ) [above (below) horizon] of h(r ) must pass through small enough vicinity
341
( O 2m x R
2m, x 2m )
O 2m x R3 | x 2m, x 2m
342
343
344
| x
O0 2m y R3 | y 2m
345
of zeros set
346
347
348
349
zero. Due to the degeneracy of (1.12), the Levi-Civit connection is not available. By
appropriate nonsmooth regularization (see section 3) we obtain an Colombeau generalized
object modeling the singular Schwarzschild metric above and below horizon, i.e.,
350
ds 2 h (r )dt 2 h (r ) 1 dr 2 r 2 d 2 ,
1
2
2
2
2
ds
h
(
r
)
dt
h
(
r
)
dr
r d ,
0,1.
351
352
353
354
l
kj
355
356
357
358
and
l
kj h
GR 3 ,
1.28
l
kj
359
l
kj h
GR 3 , :
360
361
12 g [( g )
12 ( g )[( g )
l
kj
362
l
kj
363
lm
lm
l
kj
mk , j
h , h
l
kj
] .
( g ) mj , k ( g ) kj , m ] ,
mk , j ( g ) mj , k ( g ) kj , m
1.29
364
Obviously
365
366
367
g , 0,1, i.e. ( g )ij ; k 0 . Proceeding in this manner, we obtain the nonstandard result
368
R 11
R 11
369
370
371
372
R 00
m 2m,
R 00
m 2m.
1.30
Investigating the weak limit of the angular components of the generalized Ricci tensor using
the abbreviation
373
~
(r ) sin d
374
d( x)
375
376
and let
377
( S2 m (R
) ) we denote the class of all functions (x) with compact support such that
378
379
(i) supp
380
(ii)
381
382
( x) x |
x 2m (supp ( x) x | x 2m)
~
(r ) C R . Then for any function (x) S2 m (R 3 ) we get:
383
384
1.31
385
386
387
i.e., the Schwarzschild spacetime is weakly Ricci-nonflat (the origin was excluded from our
considerations). Furthermore, the Tolman formula [3], [4] for the total energy of a static and
asymptotically flat spacetime
388
389
390
ET Trr T T Ttt g d 3 x m,
391
1.32
392
393
as it should be.
394
395
396
The paper is organized in the following way: in section II we discuss the conceptual as well
as the mathematical prerequisites. In particular we comment on geometrical matters
(differentiable structure, coordinate invariance) and recall the basic facts of nonlinear
397
398
399
400
401
402
403
404
405
406
407
408
In
409
410
physical
literature
for
many
years
the
belief
that
Schwarzschild
spacetime
r 2m. In physical literature [4], [33], [34], [35] one considers the formal change of
411
412
413
above horizon
given when r 2m by
414
415
1.
2m
t r 2m ln
416
417
1.33
418
dt
419
1.34
dr
dv .
2m
1
r
420
421
422
423
2m 2
2
2
2
2
ds 2 1
dv 2drdv r (d ) sin (d ) .
r
424
1.35
425
426
When
427
t r 2m ln1
428
r
.
2m
1.36
429
430
431
dt
432
1.37
dr
dv .
2m
1
r
433
434
435
436
2m 2
ds 2
1 dv 2drdv r 2 (d ) 2 sin 2 (d ) 2 .
r
437
1.38
438
S 2 r 0 R and
439
440
Remark 1.12. (i) Note that the metric (1.33) is defined on the manifold
obviously it is regular Lorentzian metric: its coefficients are smooth.
441
442
(iii) Due to the nondegeneracy of the metric (1.32) the Levi-Civita connection
443
444
445
{} g {}[( g {} g {} g {}]
l
kj
lm
mk , j
mj , k
kj , m
1.39
446
447
448
rs
r r r
r
1
, r s 3 s , rr s 2 , r ,
2
2r
2r
2r
r
1
r ,
r , r r r rs ,
cot ,
r
r sin 2 ,
r r rs sin 2 ,
sin cos .
449
1.40
450
451
452
(iv) In physical literature [4], [33], [34], [35] by using properties (i)-(iii) this spacetime wrongly
convicted as a rigorous mathematical extension of the Schwarzschild spacetime.
453
454
(ii)
455
456
Obviously both transformations given by Eq. (1.33) and Eq. (1.36) are singular because the
both Jacobian of these transformations is singular at r 2m :
v , r r , , and
v , r r , , .
457
t
r
458
v
r 1
r
0
r
r 2m
1
1.41
459
460
and
461
t
r
462
v
r
.
r 1
2
m
r
1
0
r
1.42
463
464
465
Remark 1.14. Note first (i) such singular transformations not allowed in conventional
Lorentzian geometry and second
466
467
(ii) both Eddington-Finkelstein metrics given by Eq. (1.35) and by Eq. (1.38) again not well
defined in any rigorous mathematical sence at r 2m.
468
Remark
469
470
471
472
473
474
1.15.
From
S r 2m R
2
consideration
above
follows
that
Schwarzschild
spacetime
475
476
477
478
Remark 1.16. From consideration above it follows that it is necessary a regularization of the
Eq. (1.34) and Eq. (1.37) on horizon. However, obviously only nonsmooth regularization via
horizon r 2m is possible. Under nonsmooth regularization (see section 3) Eq. (1.34) and
Eq. (1.37) takes the form
479
dt
480
dr
1
r
r 2m2 2
0,1
dv ,
1.43
481
482
483
and
dt
484
dr
1
r
2m r 2 2
0,1
dv ,
1.44
485
486
487
488
t
r
489
v
1
r
r
0
r
r 2m
1.45
v
1
r
r
0
r
2m r .
1.46
490
491
492
and
t
r
493
494
495
From Eq. (1.43) - Eq. (1.44) one obtains generalized Eddington-Finkelstein transformations
such as
496
dt
497
498
rdr
r 2m 2
0,1
2
dv ,
1.47
499
and
500
dt
501
rdr
2m r
0,1.
2
dv ,
1.48
502
503
504
505
t
r
0
r
t
r
0
r
r 2m2 2
1.49
506
507
and
508
509
2m r 2 2 .
1.50
510
511
At point
r 2m one obtains
512
t
r
513
1 r 1
0
1
r
r r 2 m
1.51
1 r 1
0
1
r
r r 2 m
1.52
514
515
and
516
517
518
t
r
519
520
521
522
523
ds
ds
2
524
525
526
527
1
r
1
dv 2drdv r (d ) sin (d ) .
2
1.53
It is easily to verify by using formula A.2 (see appendix) that the distributional curvature
scalar R is singular at r 2m as in the case of the distributional Schwarzschild
528
529
530
531
532
533
534
535
536
537
spacetime given by Eq. (1.28). However this is not surprising because the classical
Eddington-Finkelstein spacetime and generalized Eddington-Finkelstein specetime given by
Eq. (1.53) are essentially different geometrical objects.
538
the paper
540
541
542
543
The basic idea of classical Colombeau's theory of nonlinear generalized functions [1], [2] is
regularization by sequences (nets) of smooth functions and the use of asymptotic estimates
544
539
545
546
547
548
M is defined
549
550
551
552
553
GM E M M/NM
2. 1
554
555
vector fields on
M ):
556
E M M u | KK Mkk NN
1 , , k 1 , , k XM sup |L 1 L k u p| ON as 0
2. 2
pK
557
558
559
NM u | KK M, kk 0 qq N
1 , , k 1 , , k XM sup |L 1 L k u p| Oq as 0
2. 3
pK
560
561
562
563
564
565
u clu u NM.
566
567
2. 4
569
570
The spaces of moderate resp. negligible sequences and hence the algebra itself may be
571
572
573
568
574
The spaces of moderate resp. negligible sequences and hence the algebra itself may be
575
576
577
578
579
580
581
582
M . Generalized Numbers
583
584
585
586
587
588
589
590
and as such are not the elements in the field or . Instead, they are the representatives
of Colombeau's generalized numbers. We give the exact definition of these numbers.
Within the classical distribution theory, distributions cannot be characterized by their point
values in any way similar to classical functions. On the other hand, there is a very natural
and direct way of obtaining the point values of the elements of Colombeau's algebra: points
are simply inserted into representatives. The objects so obtained are sequences of numbers,
591
594
595
596
597
into elements of
constants:
592
593
598
K E c M/Nc M
599
600
601
(with
2. 5
K = or K = for K = or K = ), where
602
E c M r KI |nn |r | On as 0
N c M r KI |mm |r | Om as 0
2. 6
I 0, 1.
603
604
605
Generalized functions on
606
values on points in
607
608
609
610
611
612
613
representative
614
615
616
617
618
619
:
be an open subset of n . On a set
620
x I |pp 0|x | Op
x I |pp 00 0 0 |x | p , for 0 0
621
622
623
2. 7
624
x y qq 0 0 |x y | q , for 0 0
625
626
627
628
and denote by
support is
2.8
629
630
631
2.9
632
633
634
weakly, i.e.
635
w -lim0 u 0
636
637
638
2. 10
639
2. 11
u 0.
Sch
640
641
642
The
643
644
645
646
generalized tensor fields T s M is defined along the same lines using analogous
asymptotic estimates with respect to the norm induced by any Riemannian metric on the
respective fibers. However, it is more convenient to use the following algebraic description of
generalized tensor fields
647
r
G rs M GM T s M,
648
649
2. 12
T s M denotes the space of smooth tensor fields and the tensor product is taken
over the module C M . Hence generalized tensor fields are just given by classical ones
650
where
651
652
653
654
with generalized coefficient functions. Many concepts of classical tensor analysis carry over
to the generalized setting [1]-[2], in particular Lie derivatives with respect to both classical
and generalized vector fields, Lie brackets, exterior algebra, etc. Moreover, generalized
655
656
GM -modules we have
657
G rs M L M G 01 M r , G 10 M s ;GM.
658
659
660
661
field
662
663
2. 13
0, 2 -tensor
664
665
666
669
670
667
668
671
672
673
674
675
676
677
678
We
679
680
681
682
briefly
recall
the
basic
generalized
Colombeau
construction.
Colombeau
683
684
685
686
(i)
687
(ii)
u C M\ and
688
689
Definition
690
691
the quotient:
G GM, of
supergeneralized functions on M, where M, dimM n, dim n , is defined as
2.8.
The
supergeneralized
Colombeau's
algebra
GM, E M M, /NM,
692
693
2. 14
694
695
696
697
698
M\ ):
E M M, u | KK M\kk NN
1 , , k 1 , , k XM\ sup |L 1 L k u p| ON , 0
&
pK
KK Mkk NN f C M 1 , , k 1 , , k XM
L w1 L wk u
sup
L w1 L wk u f
ON , 0
2. 15
fC M
NM, u | KK M\, kk 0 qq N
1 , , k 1 , , k XM\ sup |L 1 L k u p| Oq , 0
&
pK
KK Mkk NN f C M 1 , , k 1 , , k XM 2. 16
L w1 L wk u
sup
L w1 L wk u f
Oq , 0
fC M
699
700
703
L wk is denoting the weak Lie derivative in L. Schwartz sense. In the definition the
a O appears, having the following meaning:
Landau
symbol
CC 00 0 0, 1 0 a C.
704
701
702
where
705
u clu u NM,.
706
707
2.17
709
710
The spaces of moderate resp. negligible sequences and hence the algebra itself may be
711
712
713
708
714
The spaces of moderate resp. negligible sequences and hence the algebra itself may be
715
716
717
718
719
720
GM, .
721
M . Supergeneralized Numbers
722
723
724
725
726
727
728
729
and as such are not the elements in the field or . Instead, they are the representatives
of Colombeau's generalized numbers. We give the exact definition of these numbers.
Within the classical distribution theory, distributions cannot be characterized by their point
values in any way similar to classical functions. On the other hand, there is a very natural
and direct way of obtaining the point values of the elements of Colombeau's algebra: points
are simply inserted into representatives. The objects so obtained are sequences of numbers,
730
731
732
734
K (corresponding to K R resp. C ),
N
0,1
defined as the set of moderate nets of numbers ( r K
with |r | O for
m
some N ) modulo negligible nets ( |r | O for each m ); componentwise insertion of
735
736
733
K E c M, /Nc M,
737
738
739
(with
2. 18
K = or K = for K = or K = ), where
740
E c M, r KI |nn |r | On as 0
N c M, r KI |mm |r | Om as 0
I 0, 1.
741
742
,
2. 19
745
m
nets p M\
with respect to the relation p p : d h p , p O for
746
all
747
743
744
748
Supergeneralized functions on
749
750
representative
751
752
753
754
755
:
be an open subset of n \ . On a set
756
x \ I |pp 0|x | Op
757
758
759
x \ I |pp 00 0 0 |x | p , for 0 0
2. 20
760
761
762
763
764
x y qq 0 0 |x y | q , for 0 0
and denote by
compact support is
2.21
765
766
767
2.22
768
769
770
771
774
2. 23
w -lim0 u 0
772
773
We shall often write:
775
2. 24
u 0.
Sch
776
777
778
779
780
781
782
especially the space of generalized tensor fields T s M\ - is defined along the same lines
using analogous asymptotic estimates with respect to the norm induced by any Riemannian
metric on the respective fibers. However, it is more convenient to use the following algebraic
description of generalized tensor fields
783
r
G rs M, GM, T s M\,
784
785
2. 25
T s M\ denotes the space of smooth tensor fields and the tensor product is taken
over the module C M\ . Hence generalized tensor fields are just given by classical ones
786
where
787
788
789
790
with generalized coefficient functions. Many concepts of classical tensor analysis carry over
to the generalized setting [], in particular Lie derivatives with respect to both classical and
generalized vector fields, Lie brackets, exterior algebra, etc. Moreover, generalized tensor
791
792
793
0
G rs M, L M G 1 M, r , G 10 M, s ;GM, .
2. 26
794
795
796
797
798
799
802
803
804
representative
805
(resp.
806
807
isomorphism from
800
801
g ab of g ab
V of M ) there exists a
such that for fixed (small enough) g ab g ab
GM, -linear
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
differentiable manifold
824
, 0, 1, 2, 3 .
825
At
each
point
of
space-time
827
dX p dX p with
X ; 0, 1, 2, 3 appropriately.
828
829
830
826
ds2p
831
ds2 g dx dx ;, 0, 1, 2, 3.
832
833
834
835
At each point
2. 27
ds2p dX p dX p with 1, 1, 1, 1,
836
837
2. 28
X ; 0, 1, 2, 3 appropriately.
838
839
840
841
842
2. 29
843
844
with
being the smooth Christoffel symbol. The supergeneralized nonsmooth
curvature tensor is given by
845
846
847
2. 30
848
with
849
I is
850
I 1c L G L M d 4 x,
851
852
853
854
2. 31
L G is the gravitational
855
L G 1 G .
2
856
857
858
859
Here
2. 32
G g g
860
861
862
with
2. 33
863
g R G D ,
864
865
866
2. 34
with
867
D g g
2. 35
I 1c L G L M d 4 x ,
2. 36
868
869
870
I is
871
872
873
874
875
where
876
L G 1 G .
2
877
878
879
Here
2. 37
880
G g g
881
882
883
884
with
2. 38
g R G D ,
885
886
887
2. 39
with
888
D g g
889
890
891
2. 40
892
893
894
895
R R
2. 41
R R .
2. 42
896
897
898
899
900
G R 1 R T ,
2
901
902
903
follows, where
2. 43
T is defined by
904
T
905
906
T
g
with
T 2g
907
908
909
910
911
2. 44
L M
g
2. 45
being the energy-momentum density of the classical gravity source. Thus we have defined
the supergeneralized scalar curvature by
912
913
R R
2. 46
R R .
2. 47
914
915
916
G R 1 R T ,
2
917
918
follows, where
T is defined by
T
919
920
2. 48
2. 49
L M
g
2. 50
with
2g
921
922
923
924
925
L G
t L G
g ,
g ,
926
927
with
2. 51
928
L G
L G
g ,
g ,
2. 52
g , g /x .
929
930
931
932
933
934
935
936
with
937
938
In this section we leave the neighborhood of the singularity at the origin and turn to the
singularity at the horizon. The question we are aiming at is the following: using distributional
939
940
941
geometry (thus without leaving Schwarzschild coordinates), is it possible to show that the
horizon singularity of the Schwarzschild metric is not merely only a coordinate singularity. In
order to investigate this issue we calculate the distributional curvature at horizon in
942
943
944
t, r, , , r 2m
the Schwarzschild metric (1.12) takes the form above horizon r 2m and below horizon
r 2m correspondingly
945
946
above horizon r 2m :
ds 2 h rdt 2 h r 1 dr 2 r 2 d2 ,
r 2m
h r 1 2m
r r
below horizon r 2m :
3. 1
ds 2 h rdt 2 h r 1 dr 2 r 2 d2 ,
2m r
h r 1 2m
r
r
947
948
949
950
951
952
953
Remark 3.1. Following the above discussion we consider the metric coefficients
954
r 2m : r 2m r 2m 2 2 ,
r 2m : 2m r 2m r 2 2 .
955
956
957
958
959
Remark 3.2. Note that, accordingly, we have fixed the differentiable structure of the
manifold: the usual Schwarzschild coordinates and the Cartesian coordinates associated
960
961
962
correspondingly
963
964
r 2m through the
horizon. Therefore we have above horizon r 2m and below horizon r 2m
hr
where
r r2m if r 2m
0 if r 2m
h r
h r
r 2m 2 2
h 1 r
r , r 2m
r 2m
, r 2m
h 1 r
r r2m if r 2m
h r
0 if r 2m
2m r 2 2
r
h r
3. 2
G 3 , B 0, 2m ,
where B 0, 2m x 3 |0 x 2m
965
966
967
r , r 2m
r 2m
, r 2m
r
r 2m 2 2
h 1 r
G 3 , B 0, 2m
Inserting (3.2) into (3.1) we obtain a generalized object modeling the singular Schwarzschild
metric above (below) gorizon, i.e.
1 2
ds 2
h rdt h r dr
1 2
ds 2
h rdt h r dr
968
969
970
R 00
R 22
971
972
973
r 2 d2 ,
r 2 d2
3. 3
R 11
R 33
2
1 h
r h
2
h
1 h
.
r
r2
3. 4
h
r
r 2m
r r 2m 2 2
1/2
1/2
r 2m 2 2
r2
rh
1 h
r 2m
r r 2m 2 2
r 2m
r 2m 2 2
1/2
1/2
r2
1/2
r 2m 2 2
r
1/2
1.
h r
r 2m
r r 2m 2 2
1
r r 2m 2 2
r2
1/2
r r 2m
2
r 2m
r 2m 2 2
3/2
r 2m
r 2m 2 2
2 r 2m 2 2
1/2
r2
r 2m 2
1/2
r 2m 2 2
1/2
r 2m 2 2
r 2m 2 2
r 2m
r 2m 2 2
1/2
r 2m 2 2
1/2
1/2
r3
3. 5
r 2 h
2rh
r2
1
r r 2m 2 2
r2
1/2
r 2m 2
r r 2m
2
r 2m
r 2m 2 2
r 2m
2r
r r 2m 2 2
r
r 2m 2 2
974
975
976
1/2
3/2
1/2
1/2
r2
3/2
r 2m
r 2m 2 2
1/2
1/2
2 r 2m 2 2
r
1/2
1/2
2 r 2m 2 2
1/2
rr 2m 2
r 2m 2 2
1/2
1/2
2r 2m
r
r 2m 2 2
r 2m
r 2m 2 2
r 2m 2 2
r 2m
r 2m 2 2
r3
rr 2m 2
r 2m
r 2m 2 2
2 r 2m 2 2
1/2
3/2
1/2
977
978
Investigating the weak limit of the angular components of the Ricci tensor (using the
abbreviation
r sind dx
979
980
981
982
(i) supp
983
(ii) (ii)
984
x x|x 2m
r C .
x S 2m 3 we get:
R 22
x d 3 x
R 33
x d 3 x
rh 1 h rdr
3. 6
2m
R
2m
r 2m
r 2m 2 2
1/2
rdr.
rdr
2m
985
986
987
By replacement
R xd
K
2
2
xd x
x R
K
3
3
R2m
~
u(u 2m)du
u
0
2 1/ 2
R2m
~
(u 2m)du.
3.7
988
989
By replacement
990
I 3 R 33 xd 3 x I 2 R 22 x d 3 x
K
R2m
991
992
2md
2 1
1/2
R2m
2md .
3.8
993
2m
I 3 I 2
0!
R2m
1 d
1 1/2
1!
2
R2m
1 1 d
1 1/2
R 2m
2m
R2m
1 R 2m 1
1 1 d,
1/2
1
3. 9
994
995
2m as
996
l 2m
l 1 n n ,
n!
l!
2m , 1 0 , n 1
n1
2m l0
997
998
999
with
3. 10
l d l
/d l . Equations (3.9) - (3.10) give
1000
lim I 3 lim I 2
R 2m
lim 2m
lim
0
R2m
1 1 R 2m
1 1 d
1/2
1
1001
1002
1003
1004
Thus in
S 2m BR 2m S 2m 3 D 3 ,
3. 11
1005
where
1006
w lim R 33 lim I 3 0,
0
w lim
0
1007
1008
1009
For
R 11
, R 00
R 22
3. 12
lim I 2 0.
0
we get:
1010
2 R 11
K
xd 3 x 2 R 00
xd 3 x
r 2 h
2rh rdr
3. 13
2m
R
2m
r 2m 2 2
1/2
rr 2m 2
r 2m
2
2 3/2
rdr.
1011
1012
1013
By replacement
1014
I 1 2 R 11
K
xd 3 x I2 2 R 00
K
xd 3 x
r 2 h
2rh rdr
2m
R2m
1015
1016
1017
1018
By replacement
2
u 2m u u 2m
1/2
3/2
u 2 2
u 2 2
u 2mdu.
3. 14
2 R 11
xd 3 x 2 R 00
K
xd 3 x
r 2 h
2rh rdr
2m
R2m
2m
0
R2m
2 2 2
2 2 2
R2m
R2m
2m
1/2
R2m
2md
2 1
1/2
R2m
2md
2md
2 2 2
1/2
3 2 2md
2 2 2
2 1
3/2
2 2md
2 1
3/2
3 2md
3/2
3. 15
R2m
2m
I 0 I 1 2m
0!
1!
R2m
2m
0!
1!
2
1021
1022
2m
3/2
2 1
R2m
2m
R2m
2md
1019
1020
1/2
4 3 2md
2md
2 3/2
2 2
0
R2m
2 1/2
2 2
2 2 2m
2m
2
1
d
2 1 1/2 2 1 3/2
2
1
d
2 1 1/2 2 1 3/2
R2m
2
1
d
2 1 1/2 2 1 3/2
3. 16
R2m
2
1
d,
2 1 1/2 2 1 3/2
1023
2m as
1024
l 2m
l 1 n n ,
n!
l!
2m , 1 0 , n 1
n1
2m l0
1025
1026
1027
with
3. 17
l d l
/d l . Equation (3.17) gives
1028
w - lim I 0 w - lim I 1
0
R2m
2m 2m lim
0
2m 2m lim
s
1029
1030
1031
0
s
2
1
d
2 1 1/2 2 1 3/2
s
2 d
d
0 2 1 1/2
2 1 3/2
2m 2m.
3. 18
1032
s
lim
s
1033
1034
1035
Thus in
2 d
d
2
2
3/2
1
u 1 1/2
0
3. 19
S 2m B 2m, R S 2m 3 we obtain
1036
3. 20
1041
1042
R 00
R 22
R 11
R 33
1043
1044
1045
2 h ,
1 h
2 r
h 1 h
.
r
r2
3. 21
1046
h r
2m r 2 2
r
r r2m h r
h r
h r
r 2m
r r 2m 2 2
1/2
h r, r 2m.
r 2m 2 2
r2
1/2
rh 1 h rh 1 h
r 2m
r 2m 2 2
1/2
1.
3. 22
h
r h r
r 2m
2
r r 2m 2 2
1/2
2 r 2m 2 2
1/2
r3
r 2 h
2rh r h 2rh
r
r 2m 2 2
1/2
rr 2m 2
r 2m 2 2
3/2
1047
1048
1049
1050
Investigating the weak limit of the angular components of the Ricci tensor (using the
abbreviation
1051
1052
1053
sind dx
1054
1055
where
1056
R 22 x d 3 x
R 33 x d 3 x
2m
rh 1 h rdr
0
2m
2m
r 2m
r 2m 2 2
3. 23
rdr
1/2
rdr.
0
1057
1058
1059
By replacement
1060
R 22
0
2m
1061
1062
1063
By replacement
xd 3 x
u u 2mdu
u 2 2
1/2
R 33
xd 3 x
u 2mdu.
3. 24
2m
1064
I 3 R 33 xd 3 x I 2 R 22 x d 3 x
K
2m
1065
1066
1067
1068
2md
2 1
1/2
2m
2md ,
3. 25
1069
2m
I 3 I 2
0!
2
1!
1 d
1 1/2
2m
1 1 d
1/2
1
2m
2m 1
2
2m
2m
1 2m
3. 26
1 1 d,
2
1/2
1
1070
1071
1072
2m as
1073
l 2m
l 1 n n ,
n!
l!
2m , 1 0 , n 1
n1
2m l0
1074
1075
1076
with
3. 27
l d l
/dr l . Equation (3.27) gives
1077
lim I 3 lim I 2
lim 2m 1
lim
0
1078
1079
2
2
2m
2m
1 2m
1 1 d
2
1/2
1
0.
3. 28
1080
1081
Thus in S 2m BR 2m
from Eq. (3.28) we obtain
S 2m 3 , where B 0, 2m x 3 |0 x 2m
1082
w lim R 33 lim I 3 0.
0
w lim
0
1083
1084
1085
For
R 11
, R 00
R 22
3. 29
lim I 2 0.
0
we get:
1086
2 R 11
K
xd 3 x 2 R 00
K
xd 3 x
2m
r 2 h
2rh rdr
3. 30
0
2m
r
r 2m 2 2
1/2
rr 2m 2
r 2m
2
2 3/2
rdr.
1087
1088
1089
By replacement
1090
I1 2 R 11 xd 3 x I2 2 R 00 xd 3 x
2m
r 2 h
2rh rdr
0
0
2m
1091
1092
1093
By replacement
2
u 2m u u 2m
1/2
3/2
u 2 2
u 2 2
u 2mdu.
3.31
1094
2 R 11
K
xd 3 x 2 R 00
K
xd 3 x
r 2 h
2rh rdr
2m
2m
2m
2 2 2
2m
2 2 2
2m
2 1
2m
2m
3/2
1/2
2 2 2
2 1
1/2
2 1
2m
3. 32
3/2
3 2md
1/2
3 2 2md
2m
2md
2m
2m
2md
2 2 2
2m
2md
2md
2m
1/2
4 3 2md
2 3/2
2 2
2 2md
2 1/2
2 2
2 2 2m
3/2
2 2md
2m
2 1
3/2
1095
1096
1097
1098
I 0
I 1
1!
1099
1100
1101
2m
2m l
2m
0!
1
0
2m
2
1
d
2 1 1/2 2 1 3/2
2
1
d O2 .
2 1 1/2 2 1 3/2
2m as
3. 33
1102
l 2m
l 1 n n ,
n!
l!
2m , 1 0 , n 1
n1
2m l0
1103
1104
1105
with
3. 34
l d l
/d l . Equation (3.34) gives
1106
lim I 0 lim I 1
2m lim
0
2m
0!
2m 2m lim
s0
2m
s
0
2
1
d
2 1 1/2 2 1 3/2
0
d
2 d
s 2 1 3/2
2 1 1/2
2m 2m.
3. 35
1107
1108
1109
1110
0
1.
3. 36
3. 37
lim
s
1111
1112
1113
Thus in
d
2 d
u 2 1 1/2
2 1 3/2
S B 0, 2m S 3 we obtain
1114
1115
1116
1117
1118
1119
1120
1121
t
t
Tr
Tr
r T T Tt
r T T Tt
3. 38
g d 3 x 0
1122
Thus the Tolman formula [3], [4] for the total energy of a static and asymptotically flat
1123
1124
d 3 x the coordinate
1125
ET Trr T T Ttt
1126
3.39
g d 3 x m.
1127
1128
1129
2
2
2
ds 2
h rdt 1 C rdr r d
C r
1130
1131
1132
3. 40
1 h r1 .
r 2m
1133
R R
1 h 1 h
r
2
h
r 12
r
1 h h
r
2
3. 41
1
2 4,
2m
4 r 2m 2 2
1134
1135
1136
and
1137
R R
2
h
h
2
r
1138
1139
h
4 14 2
r
r
1
r 2m 2 2
4 .
2m 4
3. 42
1140
1141
1142
1143
1144
ds 2 hrdt 2 hr 1 dr 2 r 2 d2 ,
r 2m
hr 1 2m
r r ,
h 1 r r .
r 2m
1145
1146
1147
3. 43
2m ) in the form
1148
ds 2 A rdt 2 A r 1 rdr 2 r 2 d2 ,
A r r r2m ,
A r 1 r .
r 2m
1149
1150
3. 44
1151
1152
element of
r 2m r2 2 2m.
1153
1154
1155
1156
A 1 r as an
2
ds
takes the
form
1157
2
ds
A rdt 2 A r 1 dr 2
r 2 2 2m
r
A r
r 2 d2 ,
3. 46
A r 1
1158
1159
r
r 2 2m
2
1160
2m ) in the form
1161
ds 2 A rdt 2 A r 1 dr 2 r 2 d2 ,
r .
A r 2mr r , A r 1
2m r
1162
1163
3. 47
1164
1165
element of
A 1 r as an
1166
1169
1170
3. 48
2m r 2m r2 2 .
1167
1168
Thus below Horizon ( r
form
2
ds
takes the
1171
d s 2
A rdt 2 A r 1 dr 2
2m r 2 2
r
A r
1172
1173
1174
, A r 1
r 2 d2 ,
r
2m r 2 2
3. 49
.
1175
r 2 2 2m
1179
r 2 2 2m
1
r2
r 2 2
r 2 2 2m
r
1
3
2
2 3/2
2
r
r
r r 2
1176
1177
1178
3. 50
4A 2A C
R r 2 A
r
4r
r 2 2 2m
1
r2
r 2 2
22
r
r 2 2 2m
r
1
3
2
2 3/2
2
r
r
r r 2
1180
1181
1182
3. 51
.
r 2m one obtains
1183
R
1184
1185
6 1 1 1 5
2m 2 2m 2 2m 3
2m 3 2m 2
3. 52
1186
1187
Let us introduce now the general metric which has the form [11]:
1188
3. 53
1189
1190
1191
where
1192
Ar 2 1 a
Kr
Cr 1 a
Kr
K r 2
, Br
K 2 r
,
2 r
K 2 r
1 a f 2 ,
2
Kr
r
Dr 1 a f , K r dKr/dr, f r dfr/dr,
Kr
Kr r |a|,
a 0.
1193
1194
3. 54
1195
1196
1197
1 a 0 ,
K
1198
1199
1200
1 a
K
K 2 1 a f 2 0 .
K
3. 55
x ; 0, 1, 2, 3 with
1201
1202
3.56
1203
1204
1205
ds2 g dx dx
3. 57
g 00 A , g 0 D xr , g B C x x2 .
r
3. 58
1206
1207
1208
with
g given by
1209
1210
1211
1212
1213
Ar 2
Cr
r |a|
r
Dr
1214
1215
1216
1217
1218
r
r |a|
, B
r |a| 2
,
2 r
r |a| 2
r
f r 2 ,
2
r |a|
r
3. 59
r
f r, f r dfr/dr.
r |a|
r |a| 0 :
r |a|
r |a| 1
2 r 2 |a|
3. 60
1219
1220
1221
with
1222
A r 2
r
r |a|
r |a|
r
C r
D r
1223
1224
1225
1226
, B r
r |a| 2
,
2 r
r |a| 2
r
f r 2 ,
2
r |a|
r
3. 61
r
f r, f r dfr/dr.
r |a|
1227
|a| r 0 :
|a| r 1 |a| r |a| r 2 2
3. 62
1228
1229
with
1230
A r 2
C r
|a| r
r
D r
1231
1232
|a| r 2
r
, B r
,
|a| r
2 r
|a| r 2
r
f r 2 ,
2
r
|a|
r
r
f r, f r dfr/dr.
|a| r
3. 63
1233
Remark 3.5. Finally the metric (3.57) becomes the Colombeau object of the form
1234
ds2 g dx dx
1235
1236
with
3. 64
g given by
g 00 A r, g 0 D r xr ,
1237
1238
1239
3. 65
g B r C r x x2 .
r
R in terms of
1240
r 2 2
r 2 2 |a|
r 2 2
r 2 2 |a|
9a2
2a2
5
3
r 2 2 2 r 2 r 2 2 2
9a2
2a2
5
3
r 2 2 2 r 2 r 2 2 2
1241
1242
1243
1244
3. 66
R
a 2 2
a |a|
2
9a2
2a2
5
3
a 2 2 2 a 2 a 2 2 2
7a 2 2 ,
3. 67
1245
1246
a 2 2
a 2 2 |a|
9a2
2a2
5
3
a 2 2 2 r 2 r 2 2 2
7a 2 2
1248
2
Colombeau generalized numbers R .
1249
1250
1251
as
1247
(ii) for
R R in terms of
3a
5
2
r 2 2
5
2
r |a|
2
2a2
3
r 2 r 2 2 2
3. 68
3a2
a2
2
5
3
r 2 2 2 r 2 r 2 2 2
1252
1253
1254
1255
1256
1257
1258
1259
1260
expressed as
R well
R R in terms
12a 2
R R
r 2 2 |a|
a2
1
3
r 2 2 2
4a
r |a|
2
a
3
2
r 2 2
a2
r 2 2 |a|
1261
1262
22
92
5
2
2
r r r 2 2 2
2
44
814
r 4 r 2 2 r 2 2 3
3. 69
1263
1264
1265
(ii) for
1266
12a 2
6
r 2 |a|
R R
1267
1268
3. 70
r .
1269
1270
1271
dt
r 2 2 2m
v
r
r2 2 2m
dv
r
1272
1273
1274
3.71
3.72
and
1275
dt
1276
1277
1278
2m r2 2
r
2m r 2 2
dv
r
the metric given by Eq. (3.61) - Eq. (3.64) becomes to Colombeau metric of the form
1279
2
ds 2
A r, dv 2v D2 r, dv dr B r, C 1 v , r, dr
B r, r 2 d 2 sin2 d 2 .
3. 73
1280
1281
1282
1283
1284
1285
1286
1287
1288
D, with a metric
1289
1290
1291
|g | g m 2 R 2
1292
1293
1294
dD x 12
4. 1
1295
m 2 R , 0, 1.
1296
1297
1298
4. 2
Here
1299
|g | 1/2 |g | 1/2 g
1300
1301
4. 3
explicit, the mass m should be replaced by m/. Separating out a time coordinate
x , x x 0 , x i , i 1, 2, 3 we can write the action as
1302
With
1303
1304
1305
1306
1307
dx 0 L
, L
dD1 x
4. 4
x 0 is given by
1308
x L / 0 x |h | 1/2 n x ,
1309
1310
4.5
1312
suppressed, n is the unit normal to the surface, and |h | is the determinant of the
1313
1314
1315
1311
where
x, y
1316
1317
1318
1319
1320
i D1 x, y , 0, 1.
4. 6
y
x, y f y fx for any scalar function f D , without the use
Here d
of a metric volume element. We form now a conserved bracket from two complex
Colombeau solutions to the scalar wave equation (4.2) by
D1
D1
1321
1322
1323
1324
1327
1328
1329
1330
1331
1332
, 0, 1.
4. 7
where
1325
1326
dj
j , i/ |g | 1/2 g .
4. 8
, the
and satisfy the KG equation (4.2), hence the value of the integral in (4.7) is
independent of the spacelike surface over which it is evaluated, provided the functions
vanish at spatial infinity. The generalized KG inner product satisfies the relations
1333
1334
1335
, , , 0, 1.
4. 9
1336
We define now the annihilation operator associated with a complex Colombeau solution
1337
1338
a , .
1339
1340
1341
1342
4. 10
a a .
1343
1344
1345
4. 11
1346
a , a , .
1347
1348
1349
4. 12
1350
a , a , , a , a ,
1351
1352
4. 13
1354
1355
1356
|n, 1/n!a n |
1353
Note that if
1358
1359
1360
|. If we want to construct the full Hilbert space of the field theory in curved distributional
spacetime, how can we proceed? We should find a decomposition of the space of complex
1361
1362
1363
subspace S p and its complex conjugate S p , such that all brackets between solutions from
the two subspaces vanish. That is, we must find a direct sum decomposition:
1357
1364
S S p S p
1365
1366
1367
4. 14
such that
1368
, 0, S p
1369
1370
1371
and
4. 15
, 0, , S p .
1372
1373
4. 16
1374
1375
The condition (4.15) implies that each in S p can be scaled to define its own
harmonic oscillator sub-albegra. The second condition implies, according to (4.13), that the
1376
1377
a , a a , a 0.
1378
1379
1380
1381
Given such a decompostion a total Hilbert space H for the field theory can be defined as the
space of finite norm sums of possibly infinitely many states of the form
a 1, . . . a n, |0 ,
1382
1383
4. 18
1384
where
1385
1386
1387
representation of the field operator on this Fock space is hermitian and satisfies the
canonical commutation relations in sense of Colombeau generalized function.
1388
1389
1390
1391
ds 2 f dt 2 f dr 2
1392
1393
1394
r 2 d2d2 , 0, 1,
4. 19
where
f 0, 0, 1,
2
r d3
f 0 1 r 2 0d3 ,
L
r
1395
1396
1397
1398
1399
1400
4. 20
d2d2 is the metric of the ( d 2 )-sphere, and the AdS curvature radius squared L 2
2
is related to the cosmological constant by L d 2d 1/2. The parameter r 0 is
d3
proportional to the mass M of the spacetime: M d 2A d2 r 0 /16, where A d2
where
d 2
R ,
4d 1
4. 21
G , g , 8GT , ,
T , ~x.
1401
1402
1403
1404
The time-independence and the spherical symmetry of the metric imply the canonical
decomposition
1405
rY
lm
lm,
d2/2
r
t, r, e it
1406
1407
1408
where
4. 22
1409
d2 Ylm ll d 3Ylm ,
1410
1411
1412
1413
4.23
1414
f 2
1415
1416
1417
d 2 lm, r
d lm, r
f
f
2 V lm, r lm, r
2
dr
dr
0.
4.24
r r r by the relation
1418
dr
dr
1419
1420
f r .
4. 25
1421
By using a tortoise distributional coordinate the Eq. (4.24) can be written in the form of a
1422
1423
V lm , r
d r
dr
1424
1425
2 V r r 0.
1426
1427
constant
4. 26
1428
1429
geometry.
1430
1431
1432
1433
1434
V r r 0 as r r , i.e.
1435
t, r ~expit r r , as r~r .
1436
1437
1438
1439
Note that classically nothing should leave the horizon and thus classically only ingoing
modes (corresponding to a plus sign) should be present, i.e.
1440
1441
1442
4. 27
4.28
1443
r r
1444
1445
1446
where
f r
f 1 rdr
0.
f r
ln|r r | as r~r ,
4. 29
|r r | 2i/f r
4. 30
1449
1450
1451
1452
1453
1454
1455
1456
1457
1458
ds 2 f 2,in dt 2 dx 2
asymp. past
1459
1460
5. 1
1461
where
1462
h ,ij
1463
1464
1465
1466
written as
/f ,J
, where
can be
satisfies
1467
2
2
t
1468
1469
1470
1471
,J
V ,J ,
5.2
,in is the flat Laplace operator, ,out is the Laplace operator associated
with the spatial metric h ij , and the effective potential V J is given by
where
1472
V ,J
J f ,J
f ,J
f
1 6 J ,J
f ,J
1473
1474
f ,J m 2 R
2
f ,J
K ,J ,
5. 3
1476
K ,in 0 , K ,out K ,out x the scalar curvature associated with the spatial
distributional metric h ,ij .
1477
1478
We assume now this condition: (i) the massless ( m 0 ) field with arbitrary coupling in
spacetimes which are asymptotically flat in the past and asymptotically static in the future,
1479
1480
1481
1482
1475
1483
1484
with
u
,k
and
v
,
, can be naturally defined by the requirement that they are the solutions of Eq. (5.2)
which satisfy the asymptotic conditions:
1485
u
1486
1487
,k
past
5.4
and
v
,
1488
1489
1490
where
2 1/2 e it f ,out F , x ,
future
5.5
1491
,out V ,out xF , x 2 F , x ,
F , x
1492
1493
1494
C 3
5.6
d 3 x h F , x F ,x
out
1495
1496
5.7
1497
1498
on a Cauchy surface out in the asymptotic future. Note that each F , , 0, 1 can be
chosen to be real without loss of generality. There are reasonable situations where the
1499
distributional modes
1500
v
,
v
,
fail to form a complete set of distributional normal modes. This happens whenever
1502
1503
1501
the operator
w
,
1504
w
,
F , x
e ti/12 e ti/12
2 f ,out x
future
1505
1506
1507
1508
1509
1510
5.8
,
and their complex conjugates
are necessary in order to expand an arbitrary
Colombeau solution of Eq. (5.1). As a direct consequence, at least some of the in-modes
1511
2 x
1512
1513
e 2t
2
future
F x
f ,out x
1 Oe t ,
5. 9
F x is the eigenfunction of Eq. (5.6) associated with the lowest negative eigenvalue
2
2 , is some positive constant, and is a dimensionless constant
allowed,
1514
where
1515
1516
(typically of order unity) whose exact value depends globally on the spacetime structure
u
,k
1518
1519
1520
one would expect, these wild quantum fluctuations give an important contribution to the
vacuum energy stored in the field. In fact, the expectation value of its distributional energy-
1521
1522
momentum tensor, T , x ,
dominated by this exponential growth:
on the mode
w
,
1517
whose
1523
T ,00 x
future
2 x
1 4 2 DF 2
2
m 2 f K
2
2
F
2D2 f Df 2 Di f Di F
1 6
2
f
f F
2f
1524
1525
Oe t ,
5. 10
T ,0i x
future
2 x
1526
1527
5. 11
Di f
1 6
f
Di F
1 4
F
Oe ,
T ,ij x
future
2 x
1 2
Di F Dj F
F 2
Di Dj F
F
1 4h ij 2
DF 2
m 2 f 2 K
2
F 2
Di f Dj f Di fDj F Dj f Di F
1 6
f F
f F
f 2
R,ij
h ,ij
1528
1529
1530
1531
1532
1533
1534
1535
1536
1537
1538
1539
1540
2D f Df
D f D F
k
2
f
f F
2f
2
5. 12
Oe t ,
Di is the derivative operator compatible with the distributional metric h ,ij (so that
R,ij
out D2 ),
is the associated distributional Ricci tensor so that
ij
K ,out h R,ij ,
and we have omitted the subscript out in f ,out and
where
K ,out for simplicity. The Eqs. (5.10-5.12), together with Eq. (5.9), imply that on time
1 , the vacuum fluctuations of the field should overcome any other
scales determined by
classical source of energy, therefore taking control over the evolution of the background
geometry through the semiclassical Einstein equations (in which T , is included as a
source term for the distributional Einstein tensor). We are then confronted with a startling
situation where the quantum fluctuations of a field, whose energy is usually negligible in
comparison with classical energy components, are forced by the distributional background
spacetime to play a dominant role. We are still left with the task of showing that there exist
indeed well-behaved distributional background spacetimes in which the operator
1541
1542
depends Eq (5.9). Experience from quantum mechanics tells us that this typically occurs
1543
1544
when V ,out gets sufficiently negative over a sufficiently large region. It is easy to see
from Eq. (5.3) that, except for very special geometries (as the flat one), one can generally
1545
1546
1547
e 2t
2
future
2 r
1548
r 1/2 F r
r r 2 2
5. 13
, r r
1/4
T ,00 r
future
1 4 2 DF r 2
m 2 r r 2 2
2
F 2 r
2 r
1 6
2D f r
2
r r 2 2
1/4
Df r
r r 2 2
r r , f r r r 2 2
5.14
Di f rD F
1/4
1/2
1/4
r r 2 2
1/4
1549
T ,0i r
future
2 r
Di F r
F r
1 4
Di f r
1 6
r r
2
r r ,f r r r 2
2
1/4
5.15
1/4
1550
T ,ij r
future
2 r
1 6
h ,ij
Di F Dj F
F 2
1 2
Di Dj F
F
1 4h ij 2
DF r 2
m 2 r r 2 2
F 2 r
2
Di f Dj f
r r 2 2
1/2
2D2 f
r r 2 2
1/4
Di fDj F
r r 2 2
1/4
Df 2
2 r r 2 2
r r .
1551
1552
1/2
R,ij
1/2
5.16
Dj f Di F
r r 2 2
1/2
Dk f Dk F
r r 2 2
1/4
1553
1554
T ,
T ,
is
covariantly
conserved:
f ,out f ,out x , for instance for distributional BH geometry, this implies that the
1555
1556
1557
total vacuum energy is kept constant, although it continuously flows from spatial regions
where its density is negative to spatial regions where it is positive.
1558
1559
1560
Remark 5.2. Note that the singular behavior at r r appearing in Eq. (5.13) - Eq. (5.16)
leads only to asymptotic divergences, i.e. all the quantities remain finite everywhere except
horizon.
1561
1562
1563
1564
1565
1566
1567
1568
6.1
6.2
1569
1570
1571
1572
1573
1574
1575
Special interest attaches to the short distance behaviour of the Green functions, such as
1577
1578
point
1576
1579
1580
1581
1582
1583
1584
g x, 1 R y y 1 R ; y y y
3
6
1
2
R ;
R R
v y y y y . . .
20
45
where is the Minkowski metric tensor, and the coefficients are all evaluated at
Defining now
6. 3
y 0.
x, x
1585
1586
1587
g x,
1/4
G x, x
6. 4
1588
x,x 2 n
1589
1590
dn keiky k
6.5
1591
where
1592
1593
1594
iteration to any adiabatic order. The result to adiabatic order four (i.e., four derivatives of the
metric) is
1595
1596
1597
1598
1
2
k k 2 m 2 1 k 2 m 2 R
6
2
i 1 k 2 m 2 R ;
2 6
2
2
3
k 2 m 2
1 a k 2 m 2 1 R 2 2 a
3
6
3
where
6. 6
/k ,
a
1
30
1 R 1 R 1 R
;
2
120 ;
140 ;
R 1
60
1
60
6. 7
R ,
1599
1600
1601
1602
1603
1604
Similarly, the adiabatic expansions of other Green functions can be obtained by using the
other contours in figure 3. Substituting Eq. (6.6) into Eq. (6.5) gives
x,x 2 n
dn keiky a0 x, x ; a1 x,x ; m 2
1605
1606
1607
1608
1609
a 2 x,x ;
m 2
k 2 m 2
6.8
where
a 0 x, x ; 1
6. 9
1610
a 1 x, x ;
1 R i 1 R y 1 a y y
;
6
2 6
3
a 2 x, x ; 1 1 R 2 1 a
2 6
3
1611
1612
1613
with all geometric quantities on the right-hand side of Eq. (6.10) evaluated at
6. 10
x .
1614
1615
1616
1617
1618
1619
k 0 plane C
1/2
1620
1621
k 2 m 2 i
i dse is
1625
6.11
1622
1623
1624
k 2 m 2 i
x, x
i4
n/2
idsisn/2 exp
im 2 s
x, x
x, x ;is
2is
x, x 1 y y .
2
1626
1627
1628
6. 12
x , while the function x, x ;is has the following asymptotic adiabatic expansion
The function
1629
1630
1631
1632
6.13
G x, x :
1633
G x,x
i4
n/2
1/2
x,x ;
idsis
n/2
1634
1635
1636
where
6.14
x,x
exp im s
x,x ;is
2is
2
1637
1638
1639
6.15
1640
1641
1642
g x, 1/2
orders are
x, x ;is
isj a2 x, x ;
6. 16
j0
1643
1644
a x, x ;
1645
1646
1647
j
with a 0 x, x ; 1, the other
being given by canonical recursion
relations which enable their adiabatic expansions to be obtained. The expansions (6.13) and
(6.16) are, however, only asymptotic approximations in the limit of large adiabatic parameter
1648
T.
1649
1650
If (6.16) is substituted into (6.14) the integral can be performed to give the adiabatic
expansion of the Feynman propagator in coordinate space:
1651
G x, x
4i
n/2
1/2
x, x ;
aj x, x ;
j0
2m
n2
4
2
m
6. 17
Hn2/2 2m 2 2
1652
1653
1654
1655
1656
in which, strictly, a small imaginary part i should be subtracted from . Since we have not
imposed global boundary conditions on the distributional Green function Colombeau solution
of (6.2), the expansion (6.17) does not determine the particular vacuum state in (6.1). In
1657
1658
particular, the " i " in the expansion of G x, x only ensures that (6.17) represents the
expectation value, in some set of states, of a time-ordered product of fields. Under some
1659
1660
circumstances the use of " i " in the exact representation (6.14) may give additional
information concerning the global nature of the states.
1661
1662
1663
1664
1665
1666
1667
1668
6.2. Effective action for the quantum matter fields in curved distributional
spacetime
As in classical case one can obtain Colombeau generalized quantity W , called the
effective action for the quantum matter fields in curved distributional spcetime, which, when
functionally differentiated, yields
W
2
1
g 2 g
1669
1670
1671
1672
1673
1674
6. 18
To discover the structure of W , let us return to first principles, recalling the Colombeau
path-integral quantization procedure such as developed in []. Our notation will imply a
treatment for the scalar field, but the formal manipulations are identical for fields of higher
spins. Note that the generating functional
1675
1676
Z J
1677
1678
D exp
iS m i J x xd n x
1679
1680
1681
state
1682
1683
condition
6.19
|0, in to be unstable, i.e. it can bring about the production of particles. In flat
space, in the limit J 0, no particles are produced, and one have the normalization
1684
1685
1686
1687
Z 0
D exp
iS m i J x xd n x
J0
0 |0 1.
6.20
1688
1689
1690
6. 21
1691
even in the absence of source currents J. Hence (6.19) will no longer apply. Path-integral
1692
1693
1694
1695
1696
S m in
J x as a current
density (a scalar density in the case of scalar fields). One can thus set J = 0 in (6.19) and
examine the variation of Z 0 :
1697
1698
1699
6.22
Note that
1700
S m
2
1
g 2 g
1701
1702
1703
6. 23
6. 24
Z 0
2
1
g 2 g
1704
1705
1706
T .
1707
Z 0 expiW
1708
1709
1710
and
1711
1712
1713
6. 25
6.26
1714
1
Z 0 detG x, x 2
1715
1716
1717
6. 27
where the proportionality constant is metric-independent and can be ignored. Thus we obtain
1718
W ilnZ 0 i tr ln
2
1719
1720
1721
In (6.28)
1722
an linear space
6. 28
1723
x|x x x g x, 2
1
1724
1725
1726
6. 29
1727
G x,x x| |x .
1728
1729
6.30
1730
1731
, is defined by
tr
dn xg x,
1
2
xx;
dn xg x,
1
2
x| xx; |x .
6. 31
1732
1733
1734
as
1735
dsexps
0
1736
1737
1738
6. 32
1739
x|exps |x
i4
n/2
1/2
x, x ;
x, x
exp im s
x,x ;isis n/2
2is
2
1740
1741
1742
1743
Now, assuming
6.33
dsis1 iexps
1744
Eii
6. 34
1745
where
1746
x 0
1747
1750
1751
6. 35
Eix lnx Ox
1748
1749
where
obtain
is the Euler's constant. Substituting now (6.35) into (6.34) and letting 0 we
1752
ln
dsexps is1
ln
1753
1754
6. 36
1755
1756
1757
1758
x|ln
|x
G x,x ;m 2 dm 2
m2
1759
1760
where the integral with respect to
1762
1763
we get
1766
Eq. (6.36). Returning now to the expression (6.28) for W using Eq. (6.37) and Eq.(6.31)
1761
1764
1765
6. 37
i
2
d xg x,
n
1
2
lim
xx
G x, x ;m 2 dm 2
6. 38
m2
x x one obtains
dm 2 dn xg x,
W i
2
1767
1768
1769
1770
1771
1772
1773
1774
1
2
G x, x;m 2
6. 39
Colombeau quantity W is called as the one-loop effective action. In the case of fermion
effective actions, there would be a remaining trace over spinorial indices. From Eq. (6.39) we
may define an effective Lagrangian density
L ;e ff x
by
dn xg x,
1
2
L ;eff x
6.40
L x
g x,
1775
1776
1777
1778
Note that
1
2
;eff x
i
2
lim
xx
dm 2 G x,x ;m 2
m2
6.41
1781
L x diverges at the lower end of the s integral because the /2s damping
factor in the exponent vanishes in the limit x x . (Convergence at the upper end is
2
guaranteed by the i that is implicitly added to m
in the De Witt-Schwinger
1782
DeWitt-Schwinger expansion of
1779
1780
L x are
1783
L ;div x
2 1
32
lim
xx
1/2
x, x ;
exp
ds
s3
im 2 s
x, x
2is
6. 42
a 0 x, x ; isa 1 x, x ; is 2 a 2 x, x ;
1784
1785
1787
terms in this asymptotic expansion, involving a 3 and higher, are finite in the limit x x .
1788
1789
1786
1790
matter because (6.48) is, of course, infinite. What we require is to display the divergent
1791
1792
1793
L ;eff x
1
2 4
n/2
lim
1/2
x, x ;
xx
a j x, x ;
6. 43
j0
x, x
2is
1794
1795
can be analytically continued throughout the complex plane, then we may take the x x
1796
1797
1798
limit
1799
L ;eff x
2 4
aj x; idsisj1n/2 expim 2 s
n/2
j0
2 1 4 n/2 a j x;m 2
n/2j
j0
6. 44
j n ,
2
a j x; a j x, x;.
1800
1801
L ;e ff x as (length) 4 , even
when n 4. It is therefore necessary to introduce an arbitrary mass scale and to rewrite
1802
1803
1804
Eq. (6.44) as
L ;eff x
1808
2 4
n/2
n4
aj x;m 2 42j
j0
1805
1806
1807
If
j n
2
6.45
n 4, the first three terms of Eq. (6.45) diverge because of poles in the - functions:
n
4
4
2 On 4,
nn 2 4 n
4
2 On 4,
1 n
2
4
n
2 n
2 n 2 On 4.
2
4n
1809
1810
1811
L ;divx
6. 46
, we have
1812
L ;divx
4 n/2
4m 4 a 0 x; 2m 2 a 1 x;
a 2 x;
n2
nn 2
1 1 ln m 2
n4 2
2
1813
1814
1815
1816
The functions
(6.9)-(6.10)
6. 47
.
1817
a 0 x; 1,
a 1 x; 1 R ,
6
1
a 2 x;
R x, R x, 1 R x, R x,
180
180
1
1
1
1
R x,
R 2 x, .
6 5
2 6
1818
1819
6. 48
1820
1821
1822
L ;ren x
1823
1824
12
64
idslnis
0
3 x, x;ise ism 2
is 3
6. 49
1825
Special interest attaches to field theories in distributional spacetime in which the classical
1826
action
1827
g x, 2 xg x, g x, .
1828
1829
1830
6.50
1831
S g x, S g x,
dn x
1832
1833
1834
S g x,
g x,
g x,
6.51
1835
T g x, ,
1836
1837
S g x,
1
x
gx, 2
2 x
6. 52
1838
1839
and it is clear that if the classical action is invariant under the conformal transformations
(6.50), then the classical stress-tensor is traceless. Because conformal transformations are
1840
1841
1842
essentially a rescaling of lengths at each spacetime point x, the presence of a mass and
hence a fixed length scale in the theory will always break the conformal invariance.
Therefore we are led to the massless limit of the regularization and renormalization
1843
1844
procedures used in the previous section. Although all the higher order j 2 terms in the
DeWitt-Schwinger expansion of the effective Lagrangian (6.45) are infrared divergent at
1846
n 4 as m 0, we can still use this expansion to yield the ultraviolet divergent terms
arising from j 0, 1, and 2 in the four-dimensional case. We may put m 0
1847
immediately in the
1848
for
1845
1849
1850
1851
1852
1853
1854
1855
1856
1857
2 1 4 n/2 m
n4
a 2 x, 2 n ,
2
6. 53
which must be handled carefully. Substituting for a 2 x with n from (6.48), and
rearranging terms, we may write the divergent term in the effective action arising from (6.53)
as follows
W,div
1
2 4
n/2
1858
1859
1860
n4
2 n
2
d xg x,
n
dn xg x,
1
2
1
2
a 2 x,
G x
6.54
On 4
x R x,R x, 2R x, R x, 1 R 2 x, ,
3
G x R
x,R x,
6.55
and
1 , 1 .
120
360
1864
1865
1866
2 n
2
where
1861
1862
1863
n4
2 1 4 n/2 m
6. 56
1867
T x, ren 1/28802 x 23 R x,
1868
1869
1870
G x
1/28802 R x, R x, R x,R x, R x, .
6.57
r 2m follows that
1871
1872
1873
R R
r 2m 2 2
42m 4 .
6. 58
1874
Thus for the case of the distributional Schwarzchild spesetime given by the distributional
1875
r 2m one obtains
1876
1877
1878
1879
1880
T x,ren 28802
r 2m 2 2
42m 4 .
6.59
1881
1882
1883
1884
1885
1886
1887
1888
1889
1890
1891
1892
In this paper essentially new class Colombeau solutions to Einstein field equations is
obtained. We have shown that a successful approach for dealing with curvature tensor
valued distribution is to first impose admissible the nondegeneracy conditions on the metric
1893
1894
tensor, and then take its derivatives in the sense of classical distributions in
space.
1895
1896
The distributional meaning is then equivalent to the junction condition formalism. Afterwards,
through appropriate limiting procedures, it is then possible to obtain well behaved
1897
1898
1899
1900
1901
1902
1903
1904
1905
1906
1907
1908
1909
1910
The vacuum energy density of free scalar quantum field with a distributional background
spacetime also is considered. It has been widely believed that, except in very extreme
situations, the influence of gravity on quantum fields should amount to just small, subdominant contributions. Here we argue that this belief is false by showing that there exist
well-behaved spacetime evolutions where the vacuum energy density of free quantum fields
is forced, by the very same background distributional spacetime such BHs, to become
dominant over any classical energy density component. This semiclassical gravity effect
finds its roots in the singular behavior of quantum fields on curved spacetimes. In particular
1911
1912
1913
This paper dealing with an extension of the Einstein field equations using apparatus of
contemporary generalization of the classical Lorentzian geometry named in literature
Colombeau distributional geometry, see for example [1]-[2], [5]-[7] and [14]-[15]. The
regularizations of singularities present in some solutions of the Einstein equations is an
important part of this approach. Any singularities present in some solutions of the Einstein
equations recognized only in the sense of Colombeau generalized functions [1]-[2] and not
classically.
1914
1915
1916
1917
1918
1919
1920
1921
1922
1923
S 2m 3
ACKNOWLEDGMENTS
To reviewers provided important clarifications.
1924
1925
1926
1927
1928
1929
1930
1931
1932
1933
1934
1935
1936
1937
1938
1939
1940
1941
1942
1943
1944
1945
1946
1947
1948
1949
1950
1951
1952
1953
1954
1955
1956
1957
1958
1959
1960
1961
1962
1963
1964
1965
1966
1967
1968
1969
1970
1971
1972
1973
1974
1975
1976
REFERENCES
1. Colombeau JF. New Generalized Functions and Multiplication of Distributions. North
Holland, Amsterdam: 1984.
2. Colombeau JF. Elementary Introduction to New Generalized Functions. North
Holland, Amsterdam: 1985.
3. Tolman RC. Relativity, Thermodynamics and Cosmology. Clarendon Press, Oxford:
1962.
4. Landau LD, Lifschitz EM. The Classical Theory of Fields. Pergamon Press, Oxford:
1975.
5. Vickers JA, Wilson JP. A nonlinear theory of tensor distributions. gr-qc/9807068.
6. Vickers JA, Wilson JP. Invariance of the distributional curvature of the cone under
smooth diffeomorphisms. Class. Quantum Grav. 1999; 16: 579-588.
7. Vickers JA. Nonlinear generalized functions in general relativity. In: Grosser M,
Hrmann G, Kunzinger M, Oberguggenberger M., editors. Nonlinear Theory of
Generalized Functions, Chapman & Hall/CRC Research Notes in Mathematics. 401,
275-290, eds. Chapman & Hall CRC, Boca Raton; 1999.
8. Geroch R, Traschen J. Strings and other distributional sources in general relativity.
Phys. Rev. D 1987; 36: 1017-1031.
9. Balasin H, Nachbagauer H. On the distributional nature of the energy-momentum
tensor of a black hole or what curves the Schwarzschild geometry? Class. Quant.
Grav. 1993; 10: 2271-2278.
10. Balasin H, Nachbagauer H. Distributional energy-momentum tensor of the KerrNewman space-time family. Class. Quant. Grav. 1994; 11: 1453-1461.
11. Kawai T, Sakane E. Distributional energy-momentum densities of Schwarzschild
space-time. Prog. Theor. Phys. 1997; 98: 69-86.
12. Pantoja N, Rago H. Energy-momentum tensor valued distributions for the
Schwarzschild and Reissner-Nordstrm geometries. Preprint gr-qc/9710072, 1997.
13. Pantoja N, Rago H. Distributional sources in General Relativity: Two point-like
examples revisited. Preprint gr-qc/0009053, 2000.
14. Kunzinger M., Steinbauer R. Nonlinear distributional geometry. Acta Appl.Math.
2001.
15. Kunzinger M., Steinbauer R. Generalized pseudo-Riemannian Geometry. Preprint
mathFA/0107057, 2001.
16. Grosser M, Farkas E, Kunzinger M., Steinbauer R. On the foundations of nonlinear
generalized functions I, II. Mem. Am. Math. Soc. 2001; 153: 729.
17. Grosser M, Kunzinger M., Steinbauer R, Vickers J. A global theory of nonlinear
generalized functions. Adv. Math. 2001; to appear.
18. Schwartz L. Sur l'impossibilit de la multiplication des distributions. C. R. Acad. Sci.
Paris 1954; 239: 847848.
19. Gelfand IM, Schilov GE. Generalized functions. Vol. I: Properties and operations.
Academic Press, New York, London: 1964.
20. Parker P. Distributional geometry. J. Math. Phys. 1979; 20: 1423--1426.
21. Debney G, Kerr R, Schild A. Solutions of the Einstein and Einstein-Maxwell
equations. J. Math. Phys. 1969; 10: 1842-1854.
22. Heinzle JM, Steinbauer R. Remarks on the distributional Schwarzschild geometry.
Preprint gr-qc/0112047, 2001.
23. Abramowitz M, Stegun IA, editors. Handbook of Mathematical Functions with
Formulas, Graphs, and Mathematical Tables. 9th printing. Dover, New York: 1972.
24. Bracewell R. Heaviside's Unit Step Function. In: The Fourier Transform and Its
Applications. 3rd ed. McGraw-Hill, New York: 1999, pp. 57-61.
25. Israel W. Nuovo Cimento 1966; 44 1; Nuovo Cimento 1967; 48 463.
26. Parker PE. J. Math. Phys. 1979; 20: 1423.
1977
1978
1979
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
2021
2022
2023
2024
2025
2026
2027
2028
2029
2030
APPENDIX
2031
2032
and
2033
D , 0, 1 .
2034
Let us introduce now Colombeau generalized metric which has the form
R , R ,
R, , R , R ,
in
terms
A , B , C
of
R , R R and R R , in terms
generalized functions A r , B r , C r , D r are
2036
2037
2038
of Colombeau
expressed as
2 2 A 3 B
r
B
A
1
2
B
B
A 2
2
A 2
2
R R
A B
A B
1 A B
B
2 A
A 2
2
1 A 1 A 1 A B 1 A
r A
2 A
4 A
2 A B
1 A 2 B
B
2
A
1
r
A C D2
A
B
22
2
B
A B
A
r
1
1r
2 B
B
B
A. 2
R R
A 2
2
4
A
A
1
2 A
A
A 2
2
1 A 1 A B
r A
2 A B
2
1 B 1 A C D 1
r B
2
4
A B
r
A 2
2
2
2 A 2
1
r
1
2
A
B
B
A
B
B
1 A 1 A 1 A B B 1
B
2 A
4 A
2 A B
2
A C D2
A B
12
1
2 A B
A B
r
B
B
1 1
2 B
4 B
2039
A. 1
B rr 2 d 2 sin2 d 2 .
2035
and
B
1
2 B
B
B
A B
B
1
B
2 A B
2
2053
2040
Here
2041
A rB r C r D2 r .
2042
2043
2044
A. 3
Assume that
2045
2046
2047
2048
r 1, B r 1, D r 0.
A. 4
2049
4 A 2 A C
2 A ,
r
r
R R
2
2
A A C 2 1
1 1
A A 2
2
A A ,
2
r
r
r 2
r
R
R
2
2
A 2 2 A 4 A C 2 A .
r
r
r4
2050
2051
2052
A.5