This repository has been archived by the owner on May 5, 2022. It is now read-only.
-
Notifications
You must be signed in to change notification settings - Fork 0
/
chap6.lyx
2458 lines (1895 loc) · 68.4 KB
/
chap6.lyx
1
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
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
#LyX 2.2 created this file. For more info see http://www.lyx.org/
\lyxformat 508
\begin_document
\begin_header
\save_transient_properties true
\origin unavailable
\textclass jcuthesis
\options sort&compress
\use_default_options true
\master thesis_main.lyx
\begin_modules
theorems-ams
theorems-ams-extended
theorems-sec
\end_modules
\maintain_unincluded_children false
\language australian
\language_package default
\inputencoding auto
\fontencoding global
\font_roman "default" "default"
\font_sans "default" "default"
\font_typewriter "default" "default"
\font_math "auto" "auto"
\font_default_family default
\use_non_tex_fonts false
\font_sc false
\font_osf false
\font_sf_scale 100 100
\font_tt_scale 100 100
\graphics default
\default_output_format default
\output_sync 0
\bibtex_command default
\index_command default
\paperfontsize default
\spacing single
\use_hyperref false
\papersize default
\use_geometry false
\use_package amsmath 1
\use_package amssymb 1
\use_package cancel 1
\use_package esint 1
\use_package mathdots 1
\use_package mathtools 1
\use_package mhchem 1
\use_package stackrel 1
\use_package stmaryrd 1
\use_package undertilde 1
\cite_engine natbib
\cite_engine_type numerical
\biblio_style plain
\use_bibtopic false
\use_indices false
\paperorientation portrait
\suppress_date false
\justification true
\use_refstyle 1
\index Index
\shortcut idx
\color #008000
\end_index
\secnumdepth 3
\tocdepth 3
\paragraph_separation indent
\paragraph_indentation default
\quotes_language english
\papercolumns 1
\papersides 1
\paperpagestyle default
\tracking_changes false
\output_changes false
\html_math_output 0
\html_css_as_file 0
\html_be_strict false
\end_header
\begin_body
\begin_layout Chapter
Approaching non-equilibrium: Linear Response Theory for Symmetry Improved
Effective Actions
\begin_inset CommandInset label
LatexCommand label
name "chap:chap6"
\end_inset
\end_layout
\begin_layout Section
Synopsis
\end_layout
\begin_layout Standard
Chapter
\begin_inset CommandInset ref
LatexCommand ref
reference "chap:chap2"
\end_inset
introduced
\begin_inset Formula $n$
\end_inset
PIEAs as a useful general tool for calculations in QFT which can be extended
beyond equilibrium and beyond perturbation theory.
However, chapter
\begin_inset CommandInset ref
LatexCommand ref
reference "chap:chap3"
\end_inset
demonstrated that one of the major disadvantages of
\begin_inset Formula $n$
\end_inset
PIEAs is that finite order truncations of the effective actions generically
do not respect the symmetry properties one expects from the exact theory.
A particularly promising remedy for this problem is the symmetry improvement
formalism introduced by
\begin_inset CommandInset citation
LatexCommand citet
key "Pilaftsis2013"
\end_inset
for 2PIEAs and reviewed in chapter
\begin_inset CommandInset ref
LatexCommand ref
reference "chap:chap4"
\end_inset
.
Chapter
\begin_inset CommandInset ref
LatexCommand ref
reference "chap:chap4"
\end_inset
also extended the formalism to the symmetry improvement of 3PIEA and investigat
ed the properties of this scheme in equilibrium.
Chapter
\begin_inset CommandInset ref
LatexCommand ref
reference "chap:chap5"
\end_inset
introduced the idea of softly imposed symmetry improvement as an attempt
to avoid some of the problems of the symmetry improvement formalism, though
this was also considered only in equilibrium.
This chapter focuses on the extension to non-equilibrium problems, closely
following the paper by the author
\begin_inset CommandInset citation
LatexCommand citep
key "Brown_2016"
\end_inset
.
\end_layout
\begin_layout Standard
So far symmetry improvement has only been applied to non-gauged scalar field
theories in equilibrium.
It restores Goldstone's theorem and produces physically reasonable absorptive
parts in propagators
\begin_inset CommandInset citation
LatexCommand citep
key "Pilaftsis2013"
\end_inset
and has been shown to restore the second order phase transition of the
\begin_inset Formula $\mathrm{O}\left(4\right)$
\end_inset
linear sigma model in the Hartree-Fock approximation
\begin_inset CommandInset citation
LatexCommand citep
key "Mao2014"
\end_inset
.
It has been used to study pion strings evolving in the thermal bath of
a heavy ion collision
\begin_inset CommandInset citation
LatexCommand citep
key "Lu2015"
\end_inset
(though note that the SI-2PIEA was only used to calculate an equilibrium
finite temperature effective potential; this work did not constitute a
true non-equilibrium calculation using symmetry improved effective actions).
Symmetry improvement has also been demonstrated to improve the evaluation
of the effective potential of the standard model by taming the infrared
divergences of the Higgs sector, treated as an
\begin_inset Formula $\mathrm{O}\left(4\right)$
\end_inset
scalar field theory with gauge interactions turned off
\begin_inset CommandInset citation
LatexCommand citep
key "Pilaftsis2015,Pilaftsis2015a"
\end_inset
.
\end_layout
\begin_layout Standard
There is a strong motivation to extend symmetry improvement beyond equilibrium
since one of the major reasons for using
\begin_inset Formula $n$
\end_inset
PIEAs in the first place is their ability to handle non-equilibrium situations.
\begin_inset Formula $n$
\end_inset
PIEAs give an entirely mechanical way to set up the generic initial value
problem as a
\emph on
closed
\emph default
system of integro-differential equations directly for the mean fields and
low order correlation functions, which are simply related to the handful
of physical quantities (densities, conserved currents, etc.) one is most
often interested in.
Apart from a truncation to some finite loop order these equations need
not be subject to any further approximation.
Hence, apart from the symmetry issue and issues involved in the renormalisation
process,
\begin_inset Formula $n$
\end_inset
PIEAs give potentially the most general and accurate framework available
for the computation of real time properties in quantum field theory.
Thus an extension of the symmetry improvement technique to non-equilibrium
situations is well motivated.
The ultimate goal of this program would be a tractable and manifestly gauge
invariant set of equations of motion for highly excited Yang-Mills-Higgs
theories with chiral fermion matter based on the self-consistently complete
4PIEA.
In the meantime this chapter is restricted to an analysis of the symmetry
improved 2PIEA for scalar fields in the linear response regime.
\end_layout
\begin_layout Standard
The linear response approximation is investigated rather than a generic
non-equilibrium situation for several reasons.
First, the linear response approximation is simply far more tractable than
the general non-equilibrium situation as the response functions only depend
on the equilibrium properties of the theory.
Second, linear response is widely applicable in the real world: many systems
are
\begin_inset Quotes eld
\end_inset
close enough
\begin_inset Quotes erd
\end_inset
to equilibrium for practical purposes.
Third, the linear response approximation is a nice laboratory to isolate
the novel features of symmetry improvement constraints in non-equilibrium
settings.
Finally, it is expected that any physically reasonable formalism will have
a well formed linear response approximation, though this depends on the
assumption that the exact behaviour is an analytic (or at least not too
singular) function of the external perturbation within some neighbourhood
of zero perturbation.
This is true of all quantum mechanical systems (so long as the Hamiltonian
remains bounded below), but for field theories the infinite number of degrees
of freedom may complicate the situation.
\end_layout
\begin_layout Standard
The outline of the remainder of this chapter is as follows.
In section
\begin_inset CommandInset ref
LatexCommand ref
reference "sec:Linear-Response-Theory"
\end_inset
linear response theory is reviewed in connection with 2PIEAs.
In section
\begin_inset CommandInset ref
LatexCommand ref
reference "sec:Mechanical-Analogy-to-constrained-LRT"
\end_inset
a mechanical analogy is used, very similar to the one used to justify the
d'Alembert formalism in chapter
\begin_inset CommandInset ref
LatexCommand ref
reference "chap:chap4"
\end_inset
, to illustrate the linear response procedure for constrained systems.
In section
\begin_inset CommandInset ref
LatexCommand ref
reference "sec:Symmetry-Improvement-and-LRT"
\end_inset
the consequences of the constraints for the linear response functions of
the SI-2PIEA are derived, noting that a careful treatment of the constraint
procedure requires that not just the WI, but also its derivatives, must
vanish.
Finally the results are discussed and lessons drawn in section
\begin_inset CommandInset ref
LatexCommand ref
reference "sec:LRT-Discussion"
\end_inset
.
\end_layout
\begin_layout Section
Linear response theory and
\begin_inset Formula $n$
\end_inset
PIEA
\begin_inset ERT
status open
\begin_layout Plain Layout
}{
\end_layout
\end_inset
\begin_inset CommandInset label
LatexCommand label
name "sec:Linear-Response-Theory"
\end_inset
\end_layout
\begin_layout Standard
Linear response theory studies the effect of small externally applied perturbati
ons on a system initially in equilibrium (see, e.g.,
\begin_inset CommandInset citation
LatexCommand citep
key "Kapusta2006,Rammer2007,Stefanucci2013"
\end_inset
for informative discussions).
Consider a quantum system which is subjected to an external driving potential
\begin_inset Formula $-J\left(t\right)\hat{B}\left(t\right)$
\end_inset
where
\begin_inset Formula $J\left(t\right)$
\end_inset
is a c-number function of time representing the strength of the driving
and
\begin_inset Formula $\hat{B}\left(t\right)$
\end_inset
is the interaction Hamiltonian (the reason for the name will become apparent).
If the initial state of the system is described by a density matrix
\begin_inset Formula $\rho_{0}$
\end_inset
at time
\begin_inset Formula $t_{0}$
\end_inset
, with
\begin_inset Formula $J\left(t\right)=0$
\end_inset
for
\begin_inset Formula $t\leq t_{0}$
\end_inset
, then at time
\begin_inset Formula $t>t_{0}$
\end_inset
the expectation of an operator
\begin_inset Formula $\hat{A}$
\end_inset
(in the interaction picture with respect to the external perturbation)
is:
\begin_inset Formula
\begin{align}
\left\langle \hat{A}\left(t\right)\right\rangle & =\mathrm{Tr}\left\{ \rho\left(t\right)\hat{A}\left(t\right)\right\} \nonumber \\
& =\mathrm{Tr}\left\{ U\left(t,t_{0}\right)\rho_{0}U\left(t,t_{0}\right)^{\dagger}\hat{A}\left(t\right)\right\} \nonumber \\
& =\mathrm{Tr}\left\{ \mathrm{T}\left[\mathrm{e}^{i\int_{t_{0}}^{t}J\left(\tau\right)\hat{B}\left(\tau\right)\mathrm{d}\tau}\right]\rho_{0}\tilde{\mathrm{T}}\left[\mathrm{e}^{-i\int_{t_{0}}^{t}J\left(\tau\right)\hat{B}\left(\tau\right)\mathrm{d}\tau}\right]\hat{A}\left(t\right)\right\} \nonumber \\
& =\left\langle \tilde{\hat{A}}\left(t\right)\right\rangle +i\int_{t_{0}}^{t}\left\langle \left[\hat{A}\left(t\right),\hat{B}\left(\tau\right)\right]\right\rangle J\left(\tau\right)\mathrm{d}\tau+\mathcal{O}\left(J^{2}\right),
\end{align}
\end_inset
where
\begin_inset Formula $\left\langle \tilde{\hat{A}}\left(t\right)\right\rangle $
\end_inset
denotes the expected value in the absence of perturbation.
This leads to the definition of the response function
\begin_inset Formula
\begin{equation}
\chi^{AB}\left(t-\tau\right)=i\left\langle \left[\hat{A}\left(t\right),\hat{B}\left(\tau\right)\right]\right\rangle \Theta\left(t-\tau\right),
\end{equation}
\end_inset
(which only depends on the time difference due to the equilibrium assumption
about
\begin_inset Formula $\rho_{0}$
\end_inset
) such that
\begin_inset Formula
\begin{equation}
\delta A\left(t\right)\equiv\left\langle \hat{A}\left(t\right)\right\rangle -\left\langle \tilde{\hat{A}}\left(t\right)\right\rangle =\int_{t_{0}}^{t}\chi^{AB}\left(t-\tau\right)J\left(\tau\right)\mathrm{d}\tau+\mathcal{O}\left(J^{2}\right).
\end{equation}
\end_inset
(The limits can be pushed to
\begin_inset Formula $\pm\infty$
\end_inset
thanks to the step function in
\begin_inset Formula $\chi^{AB}$
\end_inset
, and the equation becomes trivial in the Fourier domain.) The goal of linear
response theory is to compute
\begin_inset Formula $\chi^{AB}\left(t-\tau\right)$
\end_inset
for perturbations
\begin_inset Formula $\hat{B}$
\end_inset
and observables
\begin_inset Formula $\hat{A}$
\end_inset
of interest.
The condition for validity of the approximation is that the quadratic term,
which is
\begin_inset Formula
\begin{equation}
-\int_{t_{0}}^{t}\int_{t_{0}}^{\tau_{1}}J\left(\tau_{1}\right)J\left(\tau_{2}\right)\left\langle \left[\left[\hat{A}\left(t\right),\hat{B}\left(\tau_{1}\right)\right],\hat{B}\left(\tau_{2}\right)\right]\right\rangle \mathrm{d}\tau_{2}\mathrm{d}\tau_{1},
\end{equation}
\end_inset
is much smaller than the linear term, which occurs for sufficiently small
sources
\begin_inset Formula $J$
\end_inset
and times
\begin_inset Formula $t-t_{0}$
\end_inset
.
\end_layout
\begin_layout Standard
Specialising to the scalar
\begin_inset Formula $\mathrm{O}\left(N\right)$
\end_inset
field theory of the previous few chapters, one must consider the 2PIEA
\begin_inset Formula $\Gamma\left[\varphi,\Delta\right]$
\end_inset
which can be connected to the linear response theory by expanding
\begin_inset Formula $\varphi\to\tilde{\varphi}+\delta\varphi$
\end_inset
and
\begin_inset Formula $\Delta\to\tilde{\Delta}+\delta\Delta$
\end_inset
about their source-free equilibrium values
\begin_inset Formula $\tilde{\varphi}$
\end_inset
and
\begin_inset Formula $\tilde{\Delta}$
\end_inset
determined by
\begin_inset Formula $\left(\delta\Gamma/\delta\varphi\right)_{\varphi=\tilde{\varphi},\Delta=\tilde{\Delta}}=\left(\delta\Gamma/\delta\Delta\right)_{\varphi=\tilde{\varphi},\Delta=\tilde{\Delta}}=0$
\end_inset
and matching terms order by order in the sources, treating the responses
\begin_inset Formula $\delta\varphi$
\end_inset
and
\begin_inset Formula $\delta\Delta$
\end_inset
as first order, as typical of a perturbation theory analysis.
At lowest order one finds the usual 2PI equations of motion with no sources
for the equilibrium solutions and at first order one finds equations of
motion for the perturbations:
\begin_inset Formula
\begin{align}
\frac{\delta^{2}\Gamma}{\delta\varphi_{b}\delta\varphi_{a}}\delta\varphi_{b}+\frac{\delta^{2}\Gamma}{\delta\Delta_{bc}\delta\varphi_{a}}\delta\Delta_{bc} & =-J_{a}-K_{ab}\tilde{\varphi}_{b},\\
\frac{\delta^{2}\Gamma}{\delta\varphi_{c}\delta\Delta_{ab}}\delta\varphi_{c}+\frac{\delta^{2}\Gamma}{\delta\Delta_{cd}\delta\Delta_{ab}}\delta\Delta_{cd} & =-\frac{1}{2}i\hbar K_{ab},
\end{align}
\end_inset
where all derivatives on the left hand sides are evaluated at the equilibrium
values.
It is possible to eliminate the fluctuations from these equations by introducin
g the linear response functions
\begin_inset Formula $\chi_{ab}^{\phi J}$
\end_inset
,
\begin_inset Formula $\chi_{abc}^{\phi K}$
\end_inset
,
\begin_inset Formula $\chi_{abc}^{\Delta J}$
\end_inset
and
\begin_inset Formula $\chi_{abcd}^{\Delta K}$
\end_inset
:
\begin_inset Formula
\begin{align}
\delta\varphi_{a} & =\chi_{ab}^{\phi J}J_{b}+\frac{1}{2}\chi_{abc}^{\phi K}K_{bc},\label{eq:lrt-dphi-from-response-functions}\\
\delta\Delta_{ab} & =\chi_{abc}^{\Delta J}J_{c}+\frac{1}{2}\chi_{abcd}^{\Delta K}K_{cd},\label{eq:lrt-dDelta-from-response-functions}
\end{align}
\end_inset
and demanding that the resulting equations hold for any value of the sources
\begin_inset Formula $J$
\end_inset
,
\begin_inset Formula $K$
\end_inset
.
Doing this leads to the system
\begin_inset Formula
\begin{align}
\frac{\delta^{2}\Gamma}{\delta\varphi_{b}\delta\varphi_{a}}\chi_{bd}^{\phi J}+\frac{\delta^{2}\Gamma}{\delta\Delta_{bc}\delta\varphi_{a}}\chi_{bcd}^{\Delta J} & =-\delta_{ad},\label{eq:lrt-response-funtion-eq1}\\
\frac{\delta^{2}\Gamma}{\delta\varphi_{c}\delta\Delta_{ab}}\chi_{ce}^{\phi J}+\frac{\delta^{2}\Gamma}{\delta\Delta_{cd}\delta\Delta_{ab}}\chi_{cde}^{\Delta J} & =0,\label{eq:lrt-response-funtion-eq2}\\
\frac{\delta^{2}\Gamma}{\delta\varphi_{b}\delta\varphi_{a}}\chi_{bde}^{\phi K}+\frac{\delta^{2}\Gamma}{\delta\Delta_{bc}\delta\varphi_{a}}\chi_{bcde}^{\Delta K} & =-\left(\delta_{ad}\tilde{\varphi}_{e}+\delta_{ae}\tilde{\varphi}_{d}\right),\label{eq:lrt-response-funtion-eq3}\\
\frac{\delta^{2}\Gamma}{\delta\varphi_{c}\delta\Delta_{ab}}\chi_{cef}^{\phi K}+\frac{\delta^{2}\Gamma}{\delta\Delta_{cd}\delta\Delta_{ab}}\chi_{cdef}^{\Delta K} & =-\frac{1}{2}i\hbar\left(\delta_{ae}\delta_{bf}+\delta_{af}\delta_{be}\right),\label{eq:lrt-response-funtion-eq4}
\end{align}
\end_inset
where note that in the last two equations one must symmetrise the right
hand sides before removing the source
\begin_inset Formula $K$
\end_inset
(since by the symmetry of
\begin_inset Formula $K$
\end_inset
only the symmetric part contributes).
These equations determine the linear response functions entirely in terms
of the equilibrium properties of the theory (in particular, the second
derivatives of the effective action evaluated at the equilibrium solution).
Note that the last equation can be recast as a Bethe-Salpeter equation
for the
\begin_inset Formula $\chi^{\Delta K}$
\end_inset
by using
\begin_inset Formula $\Delta^{-1}=\Delta_{0}^{-1}-\Sigma+K$
\end_inset
to write
\begin_inset Formula
\begin{align}
-\Delta_{ab}^{-1}+\Delta_{0ab}^{-1}\left[\varphi\right]-\Sigma_{ab}\left[\varphi,\Delta\right] & =-\Delta_{ab}^{-1}+\left(\Delta_{0ab}^{-1}\left[\varphi\right]-\Delta_{0ab}^{-1}\left[\tilde{\varphi}\right]\right)\nonumber \\
& +\left(\Delta_{0ab}^{-1}\left[\tilde{\varphi}\right]-\Sigma_{ab}\left[\tilde{\varphi},\tilde{\Delta}\right]\right)+\left(\Sigma_{ab}\left[\tilde{\varphi},\tilde{\Delta}\right]-\Sigma_{ab}\left[\varphi,\Delta\right]\right)\nonumber \\
& =-\delta\Delta_{ab}^{-1}+\frac{\delta\Delta_{0ab}^{-1}}{\delta\varphi_{c}}\delta\varphi_{c}-\left(\frac{\delta\Sigma_{ab}}{\delta\varphi_{c}}\delta\varphi_{c}+\frac{\delta\Sigma_{ab}}{\delta\Delta_{cd}}\delta\Delta_{cd}\right),
\end{align}
\end_inset
then using the identity
\begin_inset Formula
\begin{equation}
\delta\Delta_{ab}^{-1}=-\tilde{\Delta}_{ac}^{-1}\delta\Delta_{cd}\tilde{\Delta}_{db}^{-1}+\mathcal{O}\left(J^{2},K^{2},JK\right),\label{eq:variation-of-inverse-delta}
\end{equation}
\end_inset
and the definitions of the linear response functions followed by some rearrangem
ent to give
\begin_inset Formula
\begin{align}
\chi_{abef}^{\Delta K} & =-\tilde{\Delta}_{ac}\left(\delta_{ce}\delta_{df}+\delta_{cf}\delta_{de}\right)\tilde{\Delta}_{db}-\tilde{\Delta}_{ag}\left(\frac{\delta\Delta_{0gh}^{-1}}{\delta\varphi_{c}}-\frac{\delta\Sigma_{gh}}{\delta\varphi_{c}}\right)\tilde{\Delta}_{hb}\chi_{cef}^{\phi K}\nonumber \\
& +\left(\tilde{\Delta}_{ag}\frac{\delta\Sigma_{gh}}{\delta\Delta_{cd}}\tilde{\Delta}_{hb}\right)\chi_{cdef}^{\Delta K}.
\end{align}
\end_inset
This is an equation which determines the four point kernel
\begin_inset Formula $\chi^{\Delta K}$
\end_inset
iteratively, i.e.
a Bethe-Salpeter equation with the last quantity in braces being the Bethe-Salp
eter kernel.
\end_layout
\begin_layout Section
Mechanical analogy to illustrate constrained linear response theory
\begin_inset ERT
status open
\begin_layout Plain Layout
}{
\end_layout
\end_inset
\begin_inset CommandInset label
LatexCommand label
name "sec:Mechanical-Analogy-to-constrained-LRT"
\end_inset
\end_layout
\begin_layout Standard
Here follows a brief discussion of a very simple mechanical system which
illustrates several of the unusual features of the constraint procedure
and linear response formulation which will be used.
The chief unusual feature is that the variables being perturbed are constrained
by symmetry improvement, so a
\emph on
constrained
\emph default
linear response formalism is required.
This mechanical example shows how this can be done and, in particular,
why (a) the Lagrange multiplier diverges, (b) constraints must be imposed
in the linear response approximation to begin with, and (c) secondary constrain
ts arise.
Consider a unit mass classical particle constrained to move without friction
on a circular hoop of radius
\begin_inset Formula $r$
\end_inset
in the
\begin_inset Formula $x-y$
\end_inset
plane.
Its Lagrangian is
\begin_inset Formula
\begin{align}
L & =\frac{1}{2}\dot{x}^{2}+\frac{1}{2}\dot{y}^{2}-\lambda W+j_{x}x+j_{y}y,\\
W & =\frac{1}{4}\left(x^{2}+y^{2}-r^{2}\right)^{2},
\end{align}
\end_inset
where
\begin_inset Formula $\lambda$
\end_inset
is the Lagrange multiplier and the form of the constraint
\begin_inset Formula $W=0$
\end_inset
is chosen to mimic the singular constraint procedure.
The equations of motion are
\begin_inset Formula
\begin{align}
\ddot{x} & =-\lambda\partial_{x}W+j_{x}\nonumber \\
& =-\lambda\left(x^{2}+y^{2}-r^{2}\right)x+j_{x},\\
\ddot{y} & =-\lambda\partial_{y}W+j_{y}\nonumber \\
& =-\lambda\left(x^{2}+y^{2}-r^{2}\right)y+j_{y}.
\end{align}
\end_inset
Now consider the source free case
\begin_inset Formula $j_{x}=j_{y}=0$
\end_inset
.
The constraint terms vanish unless
\begin_inset Formula $\lambda\to\infty$
\end_inset
as
\begin_inset Formula $x^{2}+y^{2}-r^{2}\to0$
\end_inset
.
Set
\begin_inset Formula $x^{2}+y^{2}-r^{2}=\eta$
\end_inset
and
\begin_inset Formula $\lambda\eta=\omega^{2}$
\end_inset
and take the limit such that
\begin_inset Formula $\omega^{2}$
\end_inset
is a constant (recall
\begin_inset Quotes eld
\end_inset
d'Alembert's principle
\begin_inset Quotes erd
\end_inset
c.f.
section
\begin_inset CommandInset ref
LatexCommand ref
reference "sec:Justifying-the-d'Alembert-Formalism"
\end_inset
).
Note that
\begin_inset Formula $W=\eta^{2}/4$
\end_inset
.
Then the equations of motion become
\begin_inset Formula
\begin{align}
\ddot{x} & =-\omega^{2}x,\\
\ddot{y} & =-\omega^{2}y,
\end{align}
\end_inset
with the solutions
\begin_inset Formula
\begin{align}
x & =r\cos\left[\omega\left(t-t_{0}\right)\right],\\
y & =r\sin\left[\omega\left(t-t_{0}\right)\right],
\end{align}
\end_inset
where
\begin_inset Formula $t_{0}$
\end_inset
and
\begin_inset Formula $\omega$
\end_inset
are determined by the initial conditions.
For simplicity the equilibrium solution is chosen as
\begin_inset Formula $x=r$
\end_inset
and
\begin_inset Formula $y=0$
\end_inset
, which determines
\begin_inset Formula $\omega=0$
\end_inset
and
\begin_inset Formula $t_{0}=0$
\end_inset
(without loss of generality).
\end_layout
\begin_layout Standard
Now turn on the sources
\begin_inset Formula $j_{x}$
\end_inset
and
\begin_inset Formula $j_{y}$
\end_inset
and investigate the linear response by setting
\begin_inset Formula $x\to\tilde{x}+\delta x$
\end_inset
,
\begin_inset Formula $y\to\tilde{y}+\delta y$
\end_inset
,
\begin_inset Formula $\lambda\to\tilde{\lambda}+\delta\lambda$
\end_inset
where the tilde variables are the source free solutions.
The variation of the constraint is
\begin_inset Formula
\begin{equation}
\delta W=\eta\left(\tilde{x}\delta x+\tilde{y}\delta y\right)\to0,
\end{equation}
\end_inset
regardless of the behaviour of
\begin_inset Formula $\delta x$
\end_inset
and
\begin_inset Formula $\delta y$
\end_inset
, so long as they are non-singular in the
\begin_inset Formula $\eta\to0$
\end_inset
limit.
However, the first order equations of motion become
\begin_inset Formula
\begin{align}
\delta\ddot{x} & =-\delta\lambda\eta\tilde{x}-2\tilde{\lambda}\left(\tilde{x}\delta x+\tilde{y}\delta y\right)\tilde{x}-\tilde{\lambda}\eta\delta x+j_{x}\nonumber \\
& =F_{x}^{\mathrm{rad}}-\omega^{2}\delta x+j_{x}=-\omega^{2}\delta x+j_{x}^{\perp},\\
\delta\ddot{y} & =-\delta\lambda\eta\tilde{y}-2\tilde{\lambda}\left(\tilde{x}\delta x+\tilde{y}\delta y\right)\tilde{y}-\tilde{\lambda}\eta\delta y+j_{y}\nonumber \\
& =F_{y}^{\mathrm{rad}}-\omega^{2}\delta y+j_{y}=-\omega^{2}\delta y+j_{y}^{\perp},
\end{align}
\end_inset
where the radial force is defined as
\begin_inset Formula $\mathbf{F}^{\mathrm{rad}}=-\left[\delta\lambda\eta+2\tilde{\lambda}\left(\tilde{x}\delta x+\tilde{y}\delta y\right)\right]\left(\tilde{x},\tilde{y}\right)$
\end_inset
, whose physical function is to balance the applied force normal to the
constraint surface, resulting in the net transverse source
\begin_inset Formula $\mathbf{j}^{\perp}$
\end_inset
.
\end_layout
\begin_layout Standard
Now notice the terms proportional to
\begin_inset Formula $\tilde{\lambda}\left(\tilde{x}\delta x+\tilde{y}\delta y\right)$
\end_inset
in the equations of motion.
In order for these terms to be well behaved in the limit
\begin_inset Formula $\tilde{\lambda}\to\infty$
\end_inset
one must have
\begin_inset Formula $\tilde{x}\delta x+\tilde{y}\delta y\to0$
\end_inset
, i.e.
the response remains within the constraint surface (to first order).
Thus the vanishing of these terms in addition to the vanishing of
\begin_inset Formula $\delta W$
\end_inset
is required to fully enforce that the response be tangential to the constraint
surface.
Also note that by examining the
\begin_inset Formula $\delta\lambda$
\end_inset
terms in the equation of motion one can identify which component of the
applied force acts normal to the constraint surface (and hence produces
no physical response).
\end_layout
\begin_layout Standard
Applying the equilibrium solution one find
\begin_inset Formula $\mathbf{F}^{\mathrm{rad}}=-\left[\delta\lambda\eta+2\tilde{\lambda}r\delta x\right]\left(r,0\right)$
\end_inset
.
For this to be well behaved as
\begin_inset Formula $\tilde{\lambda}\to\infty$
\end_inset
requires
\begin_inset Formula $\delta x=0$
\end_inset
, which also determines
\begin_inset Formula $j_{x}^{\perp}=0$
\end_inset
via the
\begin_inset Formula $\delta x$
\end_inset
equation of motion.
The
\begin_inset Formula $\delta y$
\end_inset
equation of motion is
\begin_inset Formula
\begin{align}
\delta\ddot{y} & =j_{y},
\end{align}
\end_inset
with the solution (taking into account the initial conditions
\begin_inset Formula $\delta y_{0}=\delta\dot{y}_{0}=0$
\end_inset
):
\begin_inset Formula
\begin{equation}
\delta y=\int_{0}^{t}\int_{0}^{\tau}j_{y}\left(\tau'\right)\mathrm{d}\tau'\mathrm{d}\tau=\int_{0}^{t}\left(t-\tau\right)j_{y}\left(\tau\right)\mathrm{d}\tau.
\end{equation}
\end_inset
\end_layout
\begin_layout Standard
Now compare this to the exact solution.
Substituting the ansatz
\begin_inset Formula $x=r\cos\theta\left(t\right)$
\end_inset
,
\begin_inset Formula $y=r\sin\theta\left(t\right)$
\end_inset
, the Lagrangian and equation of motion become
\begin_inset Formula
\begin{align}
L & =\frac{1}{2}r^{2}\dot{\theta}^{2}+j_{x}r\cos\theta+j_{y}r\sin\theta,\\
\ddot{\theta} & =-\frac{j_{x}}{r}\sin\theta+\frac{j_{y}}{r}\cos\theta=\frac{j_{\theta}}{r},
\end{align}
\end_inset
where
\begin_inset Formula $j_{\theta}=-j_{x}\sin\theta+j_{y}\cos\theta$
\end_inset
is the tangential component of the force.
The solution satisfying the initial conditions
\begin_inset Formula $\theta_{0}=\dot{\theta}_{0}=0$
\end_inset
is
\begin_inset Formula
\begin{equation}
\theta=\int_{0}^{t}\left(t-\tau\right)\frac{j_{\theta}\left(\tau\right)}{r}\mathrm{d}\tau.
\end{equation}