Repository navigation
Expand file tree
/
Copy pathmultipole.tex
More file actions
651 lines (630 loc) · 26.2 KB
/
Copy pathmultipole.tex
File metadata and controls
651 lines (630 loc) · 26.2 KB
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
%
% File: multipole/multipole.tex [plain TeX code]
% Last change: December 26, 2022
%
% Derivation of the expressions for optically active polarization densities,
% from the very basics of the multipole expansion.
%
% Copyright (C) 2022, Fredrik Jonsson
%
\input epsf
\input eplain
\font\ninerm=cmr9
\font\twentyrm=cmr12 at 20 truept
\font\twelvesc=cmcsc10 at 12 truept
\input amssym % to get the {\Bbb E} font (strikethrough E)
\def\document #1 {\hsize=150mm\hoffset=4.6mm\vsize=230mm\voffset=7mm
\topskip=0pt\baselineskip=12pt\parskip=0pt\leftskip=0pt\parindent=15pt
\headline={\ifnum\pageno>1\ifodd\pageno\rightheadline\else\leftheadline\fi
\else\hfill\fi}
\def\rightheadline{\tenrm{\it #1}
\hfil{\it\date}}
\def\leftheadline{\tenrm{\it\date}
\hfil{\it #1}}
\noindent~\vskip-60pt\hskip-40pt{\epsfbox{macros/UU_logo_color.eps}}
\vskip-42pt\hfill\vbox{\hbox{{\it\author}}
\hbox{{\it\date}}}\vskip 36pt
\centerline{\twelvesc #1}
\vskip 24pt\noindent}
\def\section #1 {\medskip\goodbreak\noindent{\bf #1}
\par\nobreak\smallskip\noindent}
\def\subsection #1 {\medskip\goodbreak\noindent{\it #1}
\par\nobreak\smallskip\noindent}
\def\iint{\mathop{\int\kern-8pt\int}}
\def\iiint{\mathop{\int\kern-8pt\int\kern-8pt\int}}
\def\oiint{\mathop{\int\kern-8pt\int\kern-13.2pt{\bigcirc}}}
\def\Re{\mathop{\rm Re}\nolimits} % real part
\def\Im{\mathop{\rm Im}\nolimits} % imaginary part
\def\Tr{\mathop{\rm Tr}\nolimits} % quantum mechanical trace
\def\eqq{\mathop{\vbox{\hbox{\hskip2pt?}\vskip-6pt\hbox{=}}}}
%
% Define in which way we want displayed equation numbers to appear in
% the text. In this case, it is preferred to have the equation numbers
% as one single running index, rather than the form otherwise commonly
% found format used in books, (<chapter number>.<equation number>).
%
\def\eqconstruct#1{#1}
%
% Define in which way we want displayed subequation numbers
% to appear in the text.
%
\newcount\subref
\def\eqsubreftext#1#2{%
\subref = #2 % The space stops a <number>.
\advance\subref by 96 % `a' is character code 97.
#1{\rm\char\subref}%
}
\def\date{December 26, 2022}
\def\author{Fredrik Jonsson}
\document{Multipole expansion and optically active nonlinear media}
\vskip24pt
\section{Outline}
In this analysis, the path from the classical multipole expansion of the
potentials over to the polarizability of the medium and interpretation of the
multipole moments as individual contributing terms are as follows.
\medskip
\item{1.}{Derivation of the classical multipole expansion of the potentials due
to the fixed and static distribution of charge (say, in the classical
picture of molecule or antenna. This way, we once and for all define what
the {\it multipole moments} of the distribution is, say monopole, dipole,
quadrupole, etc. This is useful once we have the externally applied field
defined, and also particularly useful when it comes to expressing the
Hamiltonian for the energy stored by the various moments when placed in an
externally applied electric field. This expansion means expressing the
inverse distance $|{\bf x}-{\bf x}'|^{-1}$ as a Taylor series in the
expressions for the static potentials
$$
\Phi({\bf x})={{1}\over{4\pi\varepsilon_0}}
\iiint_{{\Bbb R}^3}{{\rho({\bf x}')}\over{|{\bf x}-{\bf x}'|}}\,dV',
\qquad
{\bf A}({\bf x})={{\mu_0}\over{4\pi}}
\iiint_{{\Bbb R}^3}{{{\bf J}({\bf x}')}\over{|{\bf x}-{\bf x}'|}}\,dV'.
$$
where $\Phi({\bf x})$ is the scalar potential and ${\bf A}({\bf x})$ the
vector potential. Just to summarize, this is the potentials due to the
charge and current distributions $\rho({\bf x})$ and ${\bf J}({\bf x})$
as such (and {\it not} the direct potentials from any externally applied
field).}
\item{2.}{Expansion of the {\it externally applied} electromagnetic field, in
terms of a Taylor series around the charge distribution. In this case, the
starting point is the expression for stored energy,
$$
W=\iiint_{{\Bbb R}^3}\rho({\bf x})\Phi({\bf x})\,dV,
$$
in which we express the {\it externally applied} potential $\Phi({\bf x})$
as a Taylor series around a suitable center for the charge distribution
$\rho({\bf x})$. By interpreting the various-order multipole moments of the
charge distribution from the first step, we end up with a series of
{\it interaction} terms between the multipole moments and the externally
applied field.}
\item{3.}{Going from the classical picture to the quantum-mechanical
interaction with a classical (non-quantized) external field. We here move
from the classically derived expression for the stored energy over to a
picture in which we interpret the various multipole moments of the charge
distribution as {\it quantum-mechanical operators}, that is to say taking
the step
$$
\eqalign{
H&=q\Phi(0)
-{\bf p}\cdot{\bf E}(0)
-\sum_i\sum_j Q_{ij}Q_{ij}{{\partial E_j(0)}\over{\partial x_i}}
-\ldots\cr
&\hskip10pt\to\quad
\hat{H}_{\rm I}=\hat{q}\Phi(0)
-\hat{\bf p}\cdot{\bf E}(0)
-\sum_i\sum_j \hat{Q}_{ij}\hat{Q}_{ij}{{\partial E_j(0)}
\over{\partial x_i}}
-\ldots\cr
}
$$}
\medskip
\section{Series expansion of the electrostatic potential from the charge
distribution}
$$
\Phi({\bf x})={{1}\over{4\pi\varepsilon_0}}
\iiint_{{\Bbb R}^3}{{\rho({\bf x}')}\over{|{\bf x}-{\bf x}'|}}\,dV',
$$
$$
f({\bf x}')\equiv{{1}\over{|{\bf x}-{\bf x}'|}}
={{1}\over{[(x-x')^2+(y-y')^2+(z-z')^2]^{1/2}}}
$$
$$
\eqalign{
f({\bf x}')&=
f({\bf x}')\big|_{{\bf x}'={\bf 0}}
+\sum_k x'_k
{{\partial f({\bf x}')}\over{\partial x'_k}}\bigg|_{{\bf x}'={\bf 0}}
+{{1}\over{2}}\sum_j\sum_k x'_j x'_k
{{\partial^2 f({\bf x}')}
\over{\partial x'_j\partial x'_k}}\bigg|_{{\bf x}'={\bf 0}}
\cr&\hskip100pt
+{{1}\over{6}}\sum_i\sum_j\sum_k x'_i x'_j x'_k
{{\partial^3 f({\bf x}')}
\over{\partial x'_i\partial x'_j\partial x'_k}}\bigg|_{{\bf x}'={\bf 0}}
+O(|{\bf x}'|^4)\cr
}
$$
$$
\eqalign{
f({\bf x}')\big|_{{\bf x}'={\bf 0}}&={{1}\over{|{\bf x}|}}\cr
{{\partial f({\bf x}')}\over{\partial x'_k}}\bigg|_{{\bf x}'={\bf 0}}
&={{x_k}\over{|{\bf x}|^3}}\cr
{{\partial^2 f({\bf x}')}
\over{\partial x'_j\partial x'_k}}\bigg|_{{\bf x}'={\bf 0}}
&={{3 x_j x_k - \delta_{jk}|{\bf x}|^2}\over{|{\bf x}|^5}}\cr
{{\partial^3 f({\bf x}')}
\over{\partial x'_i\partial x'_j\partial x'_k}}\bigg|_{{\bf x}'={\bf 0}}
&={{15 x_i x_j x_k - 3\varepsilon_{ijk}\delta_{jk}x_k|{\bf x}|^2}
\over{|{\bf x}|^7}}\cr
}
$$
$$
\eqalign{
\Phi({\bf x})&={{1}\over{4\pi\varepsilon_0}}\bigg\{
{{1}\over{|{\bf x}|}}
\iiint_{{\Bbb R}^3}\rho({\bf x}')\,dV'
+\sum_k
{{x_k}\over{|{\bf x}|^3}}
\iiint_{{\Bbb R}^3}x'_k\rho({\bf x}')\,dV'
\cr&\hskip100pt
+{{1}\over{2}}\sum_j\sum_k
{{3 x_j x_k - \delta_{jk}|{\bf x}|^2}
\over{|{\bf x}|^5}}
\iiint_{{\Bbb R}^3} x'_j x'_k\rho({\bf x}')\,dV'
\cr&\hskip100pt
+{{1}\over{6}}\sum_i\sum_j\sum_k
{{15 x_i x_j x_k - 3\varepsilon_{ijk}\delta_{ij}x_k|{\bf x}|^2}
\over{|{\bf x}|^7}}
\iiint_{{\Bbb R}^3} x'_i x'_j x'_k\rho({\bf x}')\,dV'
\bigg\}\cr
}
$$
\section{Traceless form of the multipole moments}
\section{Series expansion of the electrostatic potential of the externally
applied field}
The electrostatically stored energy of a charge density $\rho({\bf x})$ in an
externally applied scalar electric potential $\Phi_{\rm ext}({\bf x})$ is
$$
W=\iiint_{{\Bbb R}^3}\rho({\bf x})\Phi_{\rm ext}({\bf x})\,dV
$$
The series expansion of the potential of the externally applied electric field
is
$$
\Phi_{\rm ext}({\bf x})
=\Phi_{\rm ext}({\bf 0})
+{{\partial \Phi_{\rm ext}({\bf 0})}\over{\partial x_k}} x_k
+{{1}\over{2}}{{\partial^2\Phi_{\rm ext}({\bf 0})}
\over{\partial x_j\partial x_k}} x_j x_k
+{{1}\over{6}}{{\partial^3\Phi_{\rm ext}({\bf 0})}
\over{\partial x_i\partial x_j\partial x_k}} x_i x_j x_k
+O(|{\bf x}|^4)
$$
where multiple occurrences of indices $i,j,k$ are to be interpreted as sums,
following the Einstein convention of summation, and where we should read the
short-hand notation of the derivaties evaluated at the origo as
$$
{{\partial \Phi_{\rm ext}({\bf 0})}\over{\partial x_k}}
\equiv{{\partial \Phi_{\rm ext}({\bf x})}\over{\partial x_k}}
\bigg|_{({\bf x}={\bf 0})},\qquad
{{\partial^2\Phi_{\rm ext}({\bf 0})}\over{\partial x_j\partial x_k}}
\equiv{{\partial^2\Phi_{\rm ext}({\bf x})}\over{\partial x_j\partial x_k}}
\bigg|_{({\bf x}={\bf 0})},
\qquad\hbox{etc.}
$$
Inserting the series expansion of the externally applied potential, the
resulting stored energy hence becomes
$$
W=\Phi_{\rm ext}({\bf 0})q
+p_k{{\partial \Phi_{\rm ext}({\bf 0})}\over{\partial x_k}}
+{{1}\over{2}}Q_{jk}{{\partial^2\Phi_{\rm ext}({\bf 0})}
\over{\partial x_j\partial x_k}}
+{{1}\over{6}}Q_{ijk}{{\partial^3\Phi_{\rm ext}({\bf 0})}
\over{\partial x_i\partial x_j\partial x_k}}
+O(|{\bf x}|^4)
$$
where the net charge $q$ and the components of the dipole $p_k$, quadrupole
$Q_{jk}$, and octopole $Q_{ijk}$ moments are defined by
$$
\eqalign{
q\equiv\iiint_{{\Bbb R}^3}\rho({\bf x})\,dV,\quad
p_k\equiv\iiint_{{\Bbb R}^3}x_k\rho({\bf x})\,dV,\quad
Q_{jk}\equiv\iiint_{{\Bbb R}^3}x_j x_k\rho({\bf x})\,dV,\quad
Q_{ijk}\equiv\iiint_{{\Bbb R}^3}x_i x_j x_k\rho({\bf x})\,dV.
}
$$
Since the electric field strength is defined by the negative gradient of the
scalar potential,
$$
E_k = -{{\partial\Phi_{\rm ext}}\over{\partial x_k}},
$$
the stored energy is expressed in terms of the electric field strength as
$$
W=\Phi_{\rm ext}({\bf 0})q
-p_k E_k({\bf 0})
-{{1}\over{2}}Q_{jk}{{\partial E_k({\bf 0})}\over{\partial x_j}}
-{{1}\over{6}}Q_{ijk}{{\partial^2 E_k({\bf 0})}
\over{\partial x_i\partial x_j}}
+O(|{\bf x}|^4)
$$
\section{General multipole expansion}
$$
\Phi({\bf x})=
\iiint_{{\Bbb R}^3}{{\rho({\bf x}')}\over{4\pi\varepsilon_0}}
\sum^{\infty}_{n=0}\sum^{n}_{k=0}(-2)^k{-1/2\choose n}{n\choose k}
{{|{\bf x}'|^{2(n-k)}({\bf x}'\cdot{\bf x})^k}\over{|{\bf x}|^{2n+1}}}\,dV'
\eqdef{eq:genexpansion10}
$$
\noindent{\bf Table 1.} List of terms occurring in the general multipole
expansion~\eqref{eq:genexpansion10} of the electrostatic scalar potential
due to a charge distribution $\rho({\bf x})$. Following the multipole
expansion formalism, the terms are grouped according to their power in
the distance to the observation point from the origo, of orders
$O(1/|{\bf x}|^{m+1})$ where $m=2n-k$.
\smallskip\noindent
\vbox{\offinterlineskip
\def\tableline#1#2#3#4#5#6{%
height4pt &\omit& &\omit& &\omit& &\omit& &\omit& &\omit&\cr
& & & & & &\cr
height4pt &\omit& &\omit& &\omit& &\omit& &\omit& &\omit&\cr}
\hrule
\halign{
&\vrule#&\strut\quad\hfil#\hfil\quad\cr
\tableline{\vbox{\hbox{Class for term}\vskip4pt\hbox{($2^m$-pole)}}}
{$m=2n-k$}
{$n$}{$k$}
{$\displaystyle{(-2)^k{-1/2\choose n}{n\choose k}}$}
{$\displaystyle{{{|{\bf x}'|^{2(n-k)}({\bf x}'\cdot{\bf x})^k}
\over{|{\bf x}|^{2n+1}}}}$}
\noalign{\hrule}
\tableline{Monopole}{0}{0}{0}{1}{$1/|{\bf x}|$}
\noalign{\hrule}
\tableline{Dipole}{1}{1}{1}{1}
{${{{({\bf x}'\cdot{\bf x})}/{|{\bf x}|^3}}}$}
\noalign{\hrule}
\tableline{Quadrupole}{2}{1}{0}{$-1/2$}
{${{{|{\bf x}'|^2}/{|{\bf x}|^3}}}$}
\tableline{}{}{2}{2}{3/2}
{${{{({\bf x}'\cdot{\bf x})^2}/{|{\bf x}|^5}}}$}
\noalign{\hrule}
\tableline{Octopole}{3}{2}{1}{$-3/2$}
{${{{|{\bf x}'|^2({\bf x}'\cdot{\bf x})}/{|{\bf x}|^5}}}$}
\tableline{}{}{3}{3}{$5/2$}
{${{{({\bf x}'\cdot{\bf x})^3}/{|{\bf x}|^7}}}$}
\noalign{\hrule}
\tableline{Hexadecapole}{4}{2}{0}{$3/8$}
{${{{|{\bf x}'|^4}/{|{\bf x}|^5}}}$}
\tableline{}{}{3}{2}{$-15/4$}
{${{{|{\bf x}'|^2({\bf x}'\cdot{\bf x})^2}/{|{\bf x}|^7}}}$}
\tableline{}{}{4}{4}{$35/8$}
{${{{({\bf x}'\cdot{\bf x})^4}/{|{\bf x}|^9}}}$}
\noalign{\hrule}
\tableline{Dotriacontapole}{5}{3}{1}{$15/8$}
{${{{|{\bf x}'|^4({\bf x}'\cdot{\bf x})}/{|{\bf x}|^7}}}$}
\tableline{}{}{4}{3}{$-35/4$}
{${{{|{\bf x}'|^2({\bf x}'\cdot{\bf x})^3}/{|{\bf x}|^9}}}$}
\tableline{}{}{5}{5}{$63/8$}
{${{({\bf x}'\cdot{\bf x})^5}/{|{\bf x}|^{11}}}$}
}
\hrule
}
$$
\eqalign{
\Phi({\bf x})
&={{1}\over{4\pi\varepsilon_0}}
\sum^{\infty}_{n=0}\sum^{n}_{k=0}
(-2)^k{-1/2\choose n}{n\choose k}\iiint_{{\Bbb R}^3}
{{|{\bf x}'|^{2(n-k)}({\bf x}'\cdot{\bf x})^k}\over{|{\bf x}|^{2n+1}}}
\rho({\bf x}')\,dV'\cr
&={{1}\over{4\pi\varepsilon_0}}\bigg\{
\underbrace{
{{1}\over{|{\bf x}|}}\iiint_{{\Bbb R}^3}
\rho({\bf x}')\,dV'
}_{\hbox{monopole $\sim O(1/|{\bf x}|)$}}
\cr&\hskip50pt
+\underbrace{
{{1}\over{|{\bf x}|^3}}\iiint_{{\Bbb R}^3}
({\bf x}'\cdot{\bf x})
\rho({\bf x}')\,dV'
}_{\hbox{dipole moment $\sim O(1/|{\bf x}|^2)$}}
\cr&\hskip50pt
+\underbrace{
{{1}\over{2|{\bf x}|^5}}\iiint_{{\Bbb R}^3}
\Big[3({\bf x}'\cdot{\bf x})^2-|{\bf x}'|^2|{\bf x}|^2\Big]
\rho({\bf x}')\,dV'
}_{\hbox{quadrupole moment $\sim O(1/|{\bf x}|^3)$}}
\cr&\hskip50pt
+\underbrace{
{{1}\over{2|{\bf x}|^7}}\iiint_{{\Bbb R}^3}
\Big[5({\bf x}'\cdot{\bf x})^2-3|{\bf x}'|^2|{\bf x}|^2\Big]
({\bf x}'\cdot{\bf x})
\rho({\bf x}')\,dV'
}_{\hbox{octopole moment $\sim O(1/|{\bf x}|^4)$}}
\cr&\hskip50pt
+\underbrace{
{{1}\over{8|{\bf x}|^9}}\iiint_{{\Bbb R}^3}
\Big[
35({\bf x}'\cdot{\bf x})^4
-30|{\bf x}'|^2|{\bf x}|^2({\bf x}'\cdot{\bf x})^2
+3|{\bf x}'|^4|{\bf x}|^4
\Big]
\rho({\bf x}')\,dV'
}_{\hbox{hexadecapole moment $\sim O(1/|{\bf x}|^5)$}}
\cr&\hskip50pt
+\underbrace{
{{1}\over{8|{\bf x}|^{11}}}\iiint_{{\Bbb R}^3}
\Big[
63({\bf x}'\cdot{\bf x})^4
-70|{\bf x}'|^2|{\bf x}|^2({\bf x}'\cdot{\bf x})^2
+15|{\bf x}'|^4|{\bf x}|^4
\Big]({\bf x}'\cdot{\bf x})
\rho({\bf x}')\,dV'
}_{\hbox{dotriacontapole moment $\sim O(1/|{\bf x}|^6)$}}
\cr&\hskip50pt
+O(1/|{\bf x}|^7)
\bigg\}\cr
&=\sum^{\infty}_{m=0}\Phi^{(m)}({\bf x}),\cr
}
$$
where we defined the contributing multipole moments as follows, employing the
Einstein convention of summation over repeated indices,
$$
\eqalignno{
\Phi^{(0)}({\bf x})&\equiv{{1}\over{4\pi\varepsilon_0}}
{{1}\over{|{\bf x}|}}\iiint_{{\Bbb R}^3}
\rho({\bf x}')\,dV'&\cr
\Phi^{(1)}({\bf x})&\equiv{{1}\over{4\pi\varepsilon_0}}
{{x_k}\over{|{\bf x}|^3}}\iiint_{{\Bbb R}^3}
x'_k \rho({\bf x}')\,dV'&\cr
\Phi^{(2)}({\bf x})&\equiv{{1}\over{4\pi\varepsilon_0}}
{{1}\over{2|{\bf x}|^5}}
\Big[3 x_j x_k \iiint_{{\Bbb R}^3} x'_j x'_k \rho({\bf x}')\,dV'
- x_j x_j \iiint_{{\Bbb R}^3} x'_k x'_k\ \rho({\bf x}'),dV' \Big]&\cr
&={{1}\over{4\pi\varepsilon_0}}
{{(3 x_j x_k - |{\bf x}|^2\delta_{jk})}\over{2|{\bf x}|^5}}
\iiint_{{\Bbb R}^3} x'_j x'_k \rho({\bf x}')\,dV'&\cr
}
$$
We do here avoid using $m$ as subscript of the multipole terms $\Phi^{(m)}$ of
the potential, in order not to confuse such an index with the elements of a
vector.
\section{Multipole expansion in terms of spherical harmonics -- moving to
traceless moments}
$$
\Phi({\bf x})=\sum^{\infty}_{l=0}\sum^{l}_{m=-l}
{{4\pi}\over{2l+1}}q_{lm}{{Y_{lm}(\vartheta,\varphi)}\over{r^{l+1}}}
$$
\section{Hamiltonian in the interaction picture}
Using the Einstein convention of summation, the Hamiltonian in the interaction
picture with a classical, non-quantized field description, is
$$
\hat{H}_{\rm I}(t) = -\hat{p}_{\alpha}E_{\alpha}(t)
-{{1}\over{2}}\hat{Q}_{\alpha\beta}
{{\partial E_{\alpha}(t)}\over{\partial x_{\beta}}}
-{{1}\over{6}}\hat{Q}_{\alpha\beta\gamma}
{{\partial^2 E_{\alpha}(t)}\over{\partial x_{\beta}\partial x_{\gamma}}}
-\hat{m}_{\alpha}B_{\alpha}(t)
$$
\section{Expressing solutions to Poissons equation in terms of spherical
harmonics}
The spherical harmonic functions are well-known for providing solutions to the
variable-separated form of Laplace equation, $\nabla^2\Phi=0$. However, they
are also useful as a tool when expanding the electrostatic potential from a
charge distribution $\rho({\bf x})$, for which
$\nabla^2\Phi=\rho/\varepsilon_0$.
\subsection{Solving Laplace's equation in spherical coordinates}
Starting with the Laplace operator expressed in spherical
coordinates\numberedfootnote{We here follow the ordering convention
$(r,\vartheta,\varphi)$ for the radial distance $r$, polar angle
$\vartheta\in[0,\pi]$ and azimuthal angle $\varphi\in[0,2\pi]$,
as recommended in the ISO 80000-2:2019 standard {\it Quantities
and units -- Part 2: Mathematics}, pp. 20--21. Item no. 2-17.3.}
$(r,\vartheta,\varphi)$, for $0\le\vartheta\le\pi$ and $0\le\varphi\le2\pi$,
$$
\nabla^2\Phi = {{1}\over{r^2}}{{\partial}\over{\partial r}}
\Big(r^2{{\partial\Phi}\over{\partial r}}\bigg)
+{{1}\over{r^2\sin^2\vartheta}}{{\partial^2\Phi}\over{\partial\varphi^2}}
+{{1}\over{r^2\sin\vartheta}}{{\partial}\over{\partial\vartheta}}
\Big(\sin\vartheta{{\partial\Phi}\over{\partial\vartheta}}\bigg),
\eqdef{eq:laplace10}
$$
we may for the Laplace equation $\nabla^2\Phi=0$ separate this by
$$
\Phi(r,\varphi,\vartheta) = R(r)F(\varphi)T(\vartheta)
\eqdef{eq:laplace20}
$$
into\numberedfootnote{As a reference, see for example, Morse \& Feshbach,
{\it Methods of Theoretical Physics} (McGraw--Hill, New York, 1953),
Chapter 10, {\it Solutions of Laplace's and Poisson's Equations}, p.~1264.}
$$
\eqalignno{
{{1}\over{R(r)}}{{\partial}\over{\partial r}}
\Big(r^2{{\partial R(r)}\over{\partial r}}\bigg)&=\lambda=\hbox{const.},
\eqdefn{eq:laplace30}&\eqsubdef{eq:laplace30a}\cr
{{1}\over{F(\varphi)}}{{\partial^2 F(\varphi)}\over{\partial\varphi^2}}
&=-m^2=\hbox{const.}&\eqsubdef{eq:laplace30b}\cr
{{\sin\vartheta}\over{T(\vartheta)}}
{{\partial}\over{\partial\vartheta}}
\Big(\sin\vartheta{{\partial T(\vartheta)}\over{\partial\vartheta}}\bigg)
+\lambda\sin^2\vartheta&=m^2.
&\eqsubdef{eq:laplace30c}\cr
}
$$
We here chose the sign of $-m^2$ and its ``pre-squared'' form already from the
knowledge that $F(\varphi)$ must be a periodic (non-exponential) function, and
that we know that we at some point will use the square root of the constant
when solving the second-order linear ODE. To start with,
Eq.~\eqref{eq:laplace30b} is easily integrated to yield
$$
F_m(\varphi)=C_m\exp(im\varphi)+D_m\exp(-im\varphi),
$$
where $C_m$ and $D_m$ are constants, and where we for the sake of clarity in
notation added an explicit index on $F_m(\varphi)$. Since $F_m(\varphi)$ must
be a periodic function with a maximum period of $2\pi$, this implies that $m$
must be an integer. Due to the presence of positive as well as negative
exponent in the general solution for $F_m(\varphi)$, we may without lack of
generality remove this redundancy to have the solution for the azimuthal
dependence expressed as
$$
F_m(\varphi)=C_m\exp(im\varphi),\qquad m=\ldots,-2,-1,0,1,2,\ldots
$$
Evaluation of $F_m(\varphi)$ over $0\le\varphi\le2\pi$ then immediately yields
that these solutions indeed are orthogonal,
$$
\int^{2\pi}_0 F_m(\varphi)F_{m'}(\varphi)\,\varphi
= C_m C_{m'}2\pi\delta_{mm'},
$$
though by further applying the requirement that they in addition should be
orthonormal, in which case we relieve the constants $C_m$ from any boundary
conditions, instead lifting this over to the integration constants of other
factors in the variable separation of Eq.~\eqref{eq:laplace20}, we require
that $C_m=(2\pi)^{-1/2}$, and that hence
$$
F_m(\varphi)={{1}\over{\sqrt{2\pi}}}\exp(im\varphi),
\qquad m=\ldots,-2,-1,0,1,2,\ldots
\eqdef{eq:laplace40}
$$
The requirement on $m$ being an integer then leads to Eq.~\eqref{eq:laplace30c}
taking the Sturm-Liouville form
$$
{{1}\over{\sin\vartheta}}
{{\partial}\over{\partial\vartheta}}
\Big(\sin\vartheta{{\partial T(\vartheta)}\over{\partial\vartheta}}\bigg)
+\Big(\lambda - {{m^2}\over{\sin^2\vartheta}}\Big)T(\vartheta)=0.
\eqdef{eq:laplace50}
$$
Due to the presence of $\sin\vartheta$ in the denominators, this ODE is
singular at $\vartheta=0$ and $\vartheta=\pi$, something we in just a moment
will address when it comes to constants of integration. Changing variable to
$\xi=\cos\vartheta$, with $0\le\xi\le1$, we then have
$$
{{\partial}\over{\partial\xi}}
=-{{1}\over{\sin\vartheta}}{{\partial}\over{\partial\vartheta}},
$$
and Eq.~\eqref{eq:laplace50} hence takes the form of {\it Legendre's associated}
ordinary differential equation as
$$
(1-\xi^2){{\partial^2 T(\xi)}\over{\partial\xi^2}}
-2\xi{{\partial T(\xi)}\over{\partial\xi}}
+\Big(\lambda - {{m^2}\over{1-\xi^2}}\Big)T(\xi)=0.
\eqdef{eq:laplace60}
$$
Whenever $\lambda=n(n+1)$ with $n\ge m$ is an integer, Legendre's associated
equation has the general solution $T(\xi)=aP^m_n(\xi)+bQ^m_n(\xi)$, where $a$
and $b$ are constants and where $P^m_n(\xi)$ and $Q^m_n(\xi)$ are the respective
first and second kind associated Legendre functions of degree $n$ and order $m$.
Due to the fact that $Q^m_n(\xi)$ is unbounded at $\xi=1$ and $\xi=-1$ (that is
to say, at polar angles $\vartheta=0$ and $\vartheta=\pi$, respectively), the
requirement of a bounded solution hence yields the requirement that $b=0$. The
bounded {\it orthonormal} eigenfunctions\numberedfootnote{For the computation
of the normalization factor, see for example D.~D.~Trim {\it Applied
Partial Differential Equations} (PWS-Kent, Boston, 1990), Section 8.6,
{\it Sturm-Liouville Systems and Legendre's Differential Equation.}}
of the Sturm-Loiuville system of Eq.~\eqref{eq:laplace50} are hence, after
some algebra and by explicitly indexing the eigenfunctions,
$$
T_{nm}(\vartheta)=
\sqrt{{{(2n+1)(n-m)!}\over{2(n+m)!}}}P^m_n(\cos\vartheta).
\eqdef{eq:laplace70}
$$
With this form of coefficient, it is straightforward albeit somewhat cumbersome
to verify that the orthonormality of $T_{mn}(\vartheta)$ yields
$$
\int^{\pi}_0
T_{nm}(\vartheta) T_{n'm'}(\vartheta)\sin\vartheta
\,d\vartheta
=\delta_{nn'}\delta_{mm'}
$$
Again, by imposing orthonormality on $T_{nm}(\vartheta)$, we effectively
relieve the coefficients from being used to match requirements imposed by
boundary conditions, and we have hence left this to the integration constants
which finally will appear for the radial factor $R(r)$.
Finally, using that $\lambda=n(n+1)$ from the conclusion of the Sturm-Liouville
form of the equation for $T(\vartheta)$, Eq.~\eqref{eq:laplace30a} becomes
$$
{{\partial}\over{\partial r}}
\Big(r^2{{\partial R(r)}\over{\partial r}}\bigg)-n(n+1)R(r)=0.
\eqdef{eq:laplace80}
$$
From the shape of this equation, in particular with its terms of the form
$$
r^2{{\partial^2 R}\over{\partial r^2}},\quad
r{{\partial R}\over{\partial r}},\quad\hbox{and}\quad R,
$$
we may immediately conclude that the general solution must be a polynomial in
$r$, and we can directly integrate the radial equation to yield
$$
R_n(r)=A_n r^n + B_n r^{-(n+1)},
\eqdef{eq:laplace90}
$$
where $A_n$ and $B_n$ are constants. As the solution for $R_n(r)$ here stands,
we should keep in mind that the constants $A_n$ and $B_n$ in fact implicitly
also depend on the index $m$, as we have normalized the factors $F_m(\varphi)$
and $T_{nm}(\vartheta)$. Thus, by combining the solutions for the factors
$F_m(\varphi)$, $T_{mn}(\vartheta)$ and $R_n(r)$, of Eqs.~\eqref{eq:laplace40},
\eqref{eq:laplace70} and \eqref{eq:laplace90}, respectively, into the form
$\Phi(r,\varphi,\vartheta) = R(r)F(\varphi)T(\vartheta)$ used for the variable
separation, we arrive at the solution\numberedfootnote{Notice that
J.~D.~Jackson, {\it Classical Electrodynamics} (Wiley, New York, 1975),
2nd Edn., Eqs.~(3.2) and (3.8), in the treatise on Laplace's equation
in spherical coordinates already in the separation of variables assume
an inverse radial dependence, as
$$
\Phi(r,\varphi,\vartheta) = {{R(r)}\over{r}}F(\varphi)T(\vartheta),
$$
hence providing a slightly more elegant solution as
$$
\Phi(r,\vartheta,\varphi) = \sum^{\infty}_{n=0}\sum^{n}_{m=0}
(A_{nm} r^{n+1} + B_{nm} r^{-n)})Y_{nm}(\vartheta,\varphi).
$$}
$$
\Phi(r,\vartheta,\varphi) = \sum^{\infty}_{n=0}\sum^{n}_{m=-n}
(A_{nm} r^n + B_{nm} r^{-(n+1)})Y_{nm}(\vartheta,\varphi)
\eqdef{eq:laplace100}
$$
where\numberedfootnote{We may here recall that the ``$4\pi$'' in the
denominator of the square root stems from the $(2\pi)^{-1/2}$ occurring
in the azimuthal normalization of $F_m(\varphi)$, multiplied by the ``$2$''
in the denominator of the normalization constant of $T_{mn}(\vartheta)$.}
$$
Y_{nm}(\vartheta,\varphi)\equiv
\sqrt{{{(2n+1)(n-m)!}\over{4\pi(n+m)!}}}\exp(im\varphi)P^m_n(\cos\vartheta)
\eqdef{eq:laplace110}
$$
are the orthonormal spherical harmonic functions,\numberedfootnote{Morse \&
Feshbach, {\it Methods of Theoretical Physics} (McGraw--Hill, New York,
1953), Chapter 10, {\it Solutions of Laplace's and Poisson's Equations},
p.~1264. See also Jackson ({\it ibid.}), Eq.~(3.53).}
here listed with the arguments in order $(\vartheta,\varphi)$ to conform with
the commonly used convention of notation. The orthonormal condition on
$Y_{nm}(\vartheta,\varphi)$ over the entire enclosing sphere is from the
previously stated orthonormality conditions on $T_{nm}(\vartheta)$ and
$F_m(\varphi)$ easily verified to yield
$$
\eqalign{
\int^{\pi}_0
\int^{2\pi}_0
&Y_{nm}(\vartheta,\varphi) Y^*_{n'm'}(\vartheta,\varphi)\sin\vartheta
\,d\varphi
\,d\vartheta
\cr
&=\int^{\pi}_0
\int^{2\pi}_0
T_{nm}(\vartheta)F_m(\varphi)
\bigg(
T_{n'm'}(\vartheta)F_{m'}(\varphi)
\bigg)^*
\sin\vartheta
\,d\varphi
\,d\vartheta\cr
&=\underbrace{
\int^{2\pi}_0 F_m(\varphi) F^*_{m'}(\varphi)\,d\varphi
}_{=\delta_{mm'}}
\underbrace{
\int^{\pi}_0
T_{nm}(\vartheta) T^*_{n'm'}(\vartheta) \sin\vartheta
\,d\vartheta
}_{=\delta_{nn'}\delta_{mm'}}\cr
&=\delta_{nn'}\delta_{mm'}\cr
}
$$
\bye