\(\newcommand{\footnotename}{footnote}\)
\(\def \LWRfootnote {1}\)
\(\newcommand {\footnote }[2][\LWRfootnote ]{{}^{\mathrm {#1}}}\)
\(\newcommand {\footnotemark }[1][\LWRfootnote ]{{}^{\mathrm {#1}}}\)
\(\newcommand \ensuremath [1]{#1}\)
\(\newcommand {\LWRframebox }[2][]{\fbox {#2}} \newcommand {\framebox }[1][]{\LWRframebox } \)
\(\newcommand {\setlength }[2]{}\)
\(\newcommand {\addtolength }[2]{}\)
\(\newcommand {\setcounter }[2]{}\)
\(\newcommand {\addtocounter }[2]{}\)
\(\newcommand {\cline }[1]{}\)
\(\newcommand {\directlua }[1]{\text {(directlua)}}\)
\(\newcommand {\luatexdirectlua }[1]{\text {(directlua)}}\)
\(\newcommand {\protect }{}\)
\(\def \LWRabsorbnumber #1 {}\)
\(\def \LWRabsorbquotenumber "#1 {}\)
\(\def \mathchar {\ifnextchar "\LWRabsorbquotenumber \LWRabsorbnumber }\)
\(\def \mathcode #1={\mathchar }\)
\(\let \delcode \mathcode \)
\(\let \delimiter \mathchar \)
\(\let \LWRref \ref \)
\(\renewcommand {\ref }{\ifstar \LWRref \LWRref }\)
\(\newcommand {\intertext }[1]{\text {#1}\notag \\}\)
\(\newcommand {\mathllap }[2][]{{#1#2}}\)
\(\newcommand {\mathrlap }[2][]{{#1#2}}\)
\(\newcommand {\mathclap }[2][]{{#1#2}}\)
\(\newcommand {\mathmbox }[1]{#1}\)
\(\newcommand {\clap }[1]{#1}\)
\(\newcommand {\LWRmathmakebox }[2][]{#2}\)
\(\newcommand {\mathmakebox }[1][]{\LWRmathmakebox }\)
\(\newcommand {\cramped }[2][]{{#1#2}}\)
\(\newcommand {\crampedllap }[2][]{{#1#2}}\)
\(\newcommand {\crampedrlap }[2][]{{#1#2}}\)
\(\newcommand {\crampedclap }[2][]{{#1#2}}\)
\(\newenvironment {crampedsubarray}[1]{}{}\)
\(\newcommand {\crampedsubstack }{}\)
\(\newcommand {\smashoperator }[2][]{#2\limits }\)
\(\newcommand {\adjustlimits }{}\)
\(\newcommand {\SwapAboveDisplaySkip }{}\)
\(\require {extpfeil}\)
\(\Newextarrow \xleftrightarrow {10,10}{0x2194}\)
\(\Newextarrow \xLeftarrow {10,10}{0x21d0}\)
\(\Newextarrow \xhookleftarrow {10,10}{0x21a9}\)
\(\Newextarrow \xmapsto {10,10}{0x21a6}\)
\(\Newextarrow \xRightarrow {10,10}{0x21d2}\)
\(\Newextarrow \xLeftrightarrow {10,10}{0x21d4}\)
\(\Newextarrow \xhookrightarrow {10,10}{0x21aa}\)
\(\Newextarrow \xrightharpoondown {10,10}{0x21c1}\)
\(\Newextarrow \xleftharpoondown {10,10}{0x21bd}\)
\(\Newextarrow \xrightleftharpoons {10,10}{0x21cc}\)
\(\Newextarrow \xrightharpoonup {10,10}{0x21c0}\)
\(\Newextarrow \xleftharpoonup {10,10}{0x21bc}\)
\(\Newextarrow \xleftrightharpoons {10,10}{0x21cb}\)
\(\newcommand {\LWRdounderbracket }[3]{\underset {#3}{\underline {#1}}}\)
\(\newcommand {\LWRunderbracket }[2][]{\LWRdounderbracket {#2}}\)
\(\newcommand {\underbracket }[1][]{\LWRunderbracket }\)
\(\newcommand {\LWRdooverbracket }[3]{\overset {#3}{\overline {#1}}}\)
\(\newcommand {\LWRoverbracket }[2][]{\LWRdooverbracket {#2}}\)
\(\newcommand {\overbracket }[1][]{\LWRoverbracket }\)
\(\newcommand {\LATEXunderbrace }[1]{\underbrace {#1}}\)
\(\newcommand {\LATEXoverbrace }[1]{\overbrace {#1}}\)
\(\newenvironment {matrix*}[1][]{\begin {matrix}}{\end {matrix}}\)
\(\newenvironment {pmatrix*}[1][]{\begin {pmatrix}}{\end {pmatrix}}\)
\(\newenvironment {bmatrix*}[1][]{\begin {bmatrix}}{\end {bmatrix}}\)
\(\newenvironment {Bmatrix*}[1][]{\begin {Bmatrix}}{\end {Bmatrix}}\)
\(\newenvironment {vmatrix*}[1][]{\begin {vmatrix}}{\end {vmatrix}}\)
\(\newenvironment {Vmatrix*}[1][]{\begin {Vmatrix}}{\end {Vmatrix}}\)
\(\newenvironment {smallmatrix*}[1][]{\begin {matrix}}{\end {matrix}}\)
\(\newenvironment {psmallmatrix*}[1][]{\begin {pmatrix}}{\end {pmatrix}}\)
\(\newenvironment {bsmallmatrix*}[1][]{\begin {bmatrix}}{\end {bmatrix}}\)
\(\newenvironment {Bsmallmatrix*}[1][]{\begin {Bmatrix}}{\end {Bmatrix}}\)
\(\newenvironment {vsmallmatrix*}[1][]{\begin {vmatrix}}{\end {vmatrix}}\)
\(\newenvironment {Vsmallmatrix*}[1][]{\begin {Vmatrix}}{\end {Vmatrix}}\)
\(\newenvironment {psmallmatrix}[1][]{\begin {pmatrix}}{\end {pmatrix}}\)
\(\newenvironment {bsmallmatrix}[1][]{\begin {bmatrix}}{\end {bmatrix}}\)
\(\newenvironment {Bsmallmatrix}[1][]{\begin {Bmatrix}}{\end {Bmatrix}}\)
\(\newenvironment {vsmallmatrix}[1][]{\begin {vmatrix}}{\end {vmatrix}}\)
\(\newenvironment {Vsmallmatrix}[1][]{\begin {Vmatrix}}{\end {Vmatrix}}\)
\(\newcommand {\LWRmultlined }[1][]{\begin {multline*}}\)
\(\newenvironment {multlined}[1][]{\LWRmultlined }{\end {multline*}}\)
\(\let \LWRorigshoveleft \shoveleft \)
\(\renewcommand {\shoveleft }[1][]{\LWRorigshoveleft }\)
\(\let \LWRorigshoveright \shoveright \)
\(\renewcommand {\shoveright }[1][]{\LWRorigshoveright }\)
\(\newenvironment {dcases}{\begin {cases}}{\end {cases}}\)
\(\newenvironment {dcases*}{\begin {cases}}{\end {cases}}\)
\(\newenvironment {rcases}{\begin {cases}}{\end {cases}}\)
\(\newenvironment {rcases*}{\begin {cases}}{\end {cases}}\)
\(\newenvironment {drcases}{\begin {cases}}{\end {cases}}\)
\(\newenvironment {drcases*}{\begin {cases}}{\end {cases}}\)
\(\newenvironment {cases*}{\begin {cases}}{\end {cases}}\)
\(\newcommand {\MoveEqLeft }[1][]{}\)
\(\def \LWRAboxed #1!|!{\fbox {\(#1\)}&\fbox {\(#2\)}} \newcommand {\Aboxed }[1]{\LWRAboxed #1&&!|!} \)
\( \newcommand {\LWRABLines }[1][\Updownarrow ]{#1 \notag \\}\newcommand {\ArrowBetweenLines }{\ifstar \LWRABLines \LWRABLines } \)
\(\newcommand {\shortintertext }[1]{\text {#1}\notag \\}\)
\(\newcommand {\vdotswithin }[1]{\hspace {.5em}\vdots }\)
\(\newcommand {\LWRshortvdotswithinstar }[1]{\vdots \hspace {.5em} & \\}\)
\(\newcommand {\LWRshortvdotswithinnostar }[1]{& \hspace {.5em}\vdots \\}\)
\(\newcommand {\shortvdotswithin }{\ifstar \LWRshortvdotswithinstar \LWRshortvdotswithinnostar }\)
\(\newcommand {\MTFlushSpaceAbove }{}\)
\(\newcommand {\MTFlushSpaceBelow }{\\}\)
\(\newcommand \lparen {(}\)
\(\newcommand \rparen {)}\)
\(\newcommand {\ordinarycolon }{:}\)
\(\newcommand {\vcentcolon }{\mathrel {\mathop \ordinarycolon }}\)
\(\newcommand \dblcolon {\vcentcolon \vcentcolon }\)
\(\newcommand \coloneqq {\vcentcolon =}\)
\(\newcommand \Coloneqq {\dblcolon =}\)
\(\newcommand \coloneq {\vcentcolon {-}}\)
\(\newcommand \Coloneq {\dblcolon {-}}\)
\(\newcommand \eqqcolon {=\vcentcolon }\)
\(\newcommand \Eqqcolon {=\dblcolon }\)
\(\newcommand \eqcolon {\mathrel {-}\vcentcolon }\)
\(\newcommand \Eqcolon {\mathrel {-}\dblcolon }\)
\(\newcommand \colonapprox {\vcentcolon \approx }\)
\(\newcommand \Colonapprox {\dblcolon \approx }\)
\(\newcommand \colonsim {\vcentcolon \sim }\)
\(\newcommand \Colonsim {\dblcolon \sim }\)
\(\newcommand {\nuparrow }{\mathrel {\cancel {\uparrow }}}\)
\(\newcommand {\ndownarrow }{\mathrel {\cancel {\downarrow }}}\)
\(\newcommand {\bigtimes }{\mathop {\Large \times }\limits }\)
\(\newcommand {\prescript }[3]{{}^{#1}_{#2}#3}\)
\(\newenvironment {lgathered}{\begin {gathered}}{\end {gathered}}\)
\(\newenvironment {rgathered}{\begin {gathered}}{\end {gathered}}\)
\(\newcommand {\splitfrac }[2]{{}^{#1}_{#2}}\)
\(\let \splitdfrac \splitfrac \)
\(\newcommand {\LWRoverlaysymbols }[2]{\mathord {\smash {\mathop {#2\strut }\limits ^{\smash {\lower 3ex{#1}}}}\strut }}\)
\(\newcommand{\alphaup}{\unicode{x03B1}}\)
\(\newcommand{\betaup}{\unicode{x03B2}}\)
\(\newcommand{\gammaup}{\unicode{x03B3}}\)
\(\newcommand{\digammaup}{\unicode{x03DD}}\)
\(\newcommand{\deltaup}{\unicode{x03B4}}\)
\(\newcommand{\epsilonup}{\unicode{x03F5}}\)
\(\newcommand{\varepsilonup}{\unicode{x03B5}}\)
\(\newcommand{\zetaup}{\unicode{x03B6}}\)
\(\newcommand{\etaup}{\unicode{x03B7}}\)
\(\newcommand{\thetaup}{\unicode{x03B8}}\)
\(\newcommand{\varthetaup}{\unicode{x03D1}}\)
\(\newcommand{\iotaup}{\unicode{x03B9}}\)
\(\newcommand{\kappaup}{\unicode{x03BA}}\)
\(\newcommand{\varkappaup}{\unicode{x03F0}}\)
\(\newcommand{\lambdaup}{\unicode{x03BB}}\)
\(\newcommand{\muup}{\unicode{x03BC}}\)
\(\newcommand{\nuup}{\unicode{x03BD}}\)
\(\newcommand{\xiup}{\unicode{x03BE}}\)
\(\newcommand{\omicronup}{\unicode{x03BF}}\)
\(\newcommand{\piup}{\unicode{x03C0}}\)
\(\newcommand{\varpiup}{\unicode{x03D6}}\)
\(\newcommand{\rhoup}{\unicode{x03C1}}\)
\(\newcommand{\varrhoup}{\unicode{x03F1}}\)
\(\newcommand{\sigmaup}{\unicode{x03C3}}\)
\(\newcommand{\varsigmaup}{\unicode{x03C2}}\)
\(\newcommand{\tauup}{\unicode{x03C4}}\)
\(\newcommand{\upsilonup}{\unicode{x03C5}}\)
\(\newcommand{\phiup}{\unicode{x03D5}}\)
\(\newcommand{\varphiup}{\unicode{x03C6}}\)
\(\newcommand{\chiup}{\unicode{x03C7}}\)
\(\newcommand{\psiup}{\unicode{x03C8}}\)
\(\newcommand{\omegaup}{\unicode{x03C9}}\)
\(\newcommand{\Alphaup}{\unicode{x0391}}\)
\(\newcommand{\Betaup}{\unicode{x0392}}\)
\(\newcommand{\Gammaup}{\unicode{x0393}}\)
\(\newcommand{\Digammaup}{\unicode{x03DC}}\)
\(\newcommand{\Deltaup}{\unicode{x0394}}\)
\(\newcommand{\Epsilonup}{\unicode{x0395}}\)
\(\newcommand{\Zetaup}{\unicode{x0396}}\)
\(\newcommand{\Etaup}{\unicode{x0397}}\)
\(\newcommand{\Thetaup}{\unicode{x0398}}\)
\(\newcommand{\Varthetaup}{\unicode{x03F4}}\)
\(\newcommand{\Iotaup}{\unicode{x0399}}\)
\(\newcommand{\Kappaup}{\unicode{x039A}}\)
\(\newcommand{\Lambdaup}{\unicode{x039B}}\)
\(\newcommand{\Muup}{\unicode{x039C}}\)
\(\newcommand{\Nuup}{\unicode{x039D}}\)
\(\newcommand{\Xiup}{\unicode{x039E}}\)
\(\newcommand{\Omicronup}{\unicode{x039F}}\)
\(\newcommand{\Piup}{\unicode{x03A0}}\)
\(\newcommand{\Varpiup}{\unicode{x03D6}}\)
\(\newcommand{\Rhoup}{\unicode{x03A1}}\)
\(\newcommand{\Sigmaup}{\unicode{x03A3}}\)
\(\newcommand{\Tauup}{\unicode{x03A4}}\)
\(\newcommand{\Upsilonup}{\unicode{x03A5}}\)
\(\newcommand{\Phiup}{\unicode{x03A6}}\)
\(\newcommand{\Chiup}{\unicode{x03A7}}\)
\(\newcommand{\Psiup}{\unicode{x03A8}}\)
\(\newcommand{\Omegaup}{\unicode{x03A9}}\)
\(\newcommand{\alphait}{\unicode{x1D6FC}}\)
\(\newcommand{\betait}{\unicode{x1D6FD}}\)
\(\newcommand{\gammait}{\unicode{x1D6FE}}\)
\(\newcommand{\digammait}{\mathit{\unicode{x03DD}}}\)
\(\newcommand{\deltait}{\unicode{x1D6FF}}\)
\(\newcommand{\epsilonit}{\unicode{x1D716}}\)
\(\newcommand{\varepsilonit}{\unicode{x1D700}}\)
\(\newcommand{\zetait}{\unicode{x1D701}}\)
\(\newcommand{\etait}{\unicode{x1D702}}\)
\(\newcommand{\thetait}{\unicode{x1D703}}\)
\(\newcommand{\varthetait}{\unicode{x1D717}}\)
\(\newcommand{\iotait}{\unicode{x1D704}}\)
\(\newcommand{\kappait}{\unicode{x1D705}}\)
\(\newcommand{\varkappait}{\unicode{x1D718}}\)
\(\newcommand{\lambdait}{\unicode{x1D706}}\)
\(\newcommand{\muit}{\unicode{x1D707}}\)
\(\newcommand{\nuit}{\unicode{x1D708}}\)
\(\newcommand{\xiit}{\unicode{x1D709}}\)
\(\newcommand{\omicronit}{\unicode{x1D70A}}\)
\(\newcommand{\piit}{\unicode{x1D70B}}\)
\(\newcommand{\varpiit}{\unicode{x1D71B}}\)
\(\newcommand{\rhoit}{\unicode{x1D70C}}\)
\(\newcommand{\varrhoit}{\unicode{x1D71A}}\)
\(\newcommand{\sigmait}{\unicode{x1D70E}}\)
\(\newcommand{\varsigmait}{\unicode{x1D70D}}\)
\(\newcommand{\tauit}{\unicode{x1D70F}}\)
\(\newcommand{\upsilonit}{\unicode{x1D710}}\)
\(\newcommand{\phiit}{\unicode{x1D719}}\)
\(\newcommand{\varphiit}{\unicode{x1D711}}\)
\(\newcommand{\chiit}{\unicode{x1D712}}\)
\(\newcommand{\psiit}{\unicode{x1D713}}\)
\(\newcommand{\omegait}{\unicode{x1D714}}\)
\(\newcommand{\Alphait}{\unicode{x1D6E2}}\)
\(\newcommand{\Betait}{\unicode{x1D6E3}}\)
\(\newcommand{\Gammait}{\unicode{x1D6E4}}\)
\(\newcommand{\Digammait}{\mathit{\unicode{x03DC}}}\)
\(\newcommand{\Deltait}{\unicode{x1D6E5}}\)
\(\newcommand{\Epsilonit}{\unicode{x1D6E6}}\)
\(\newcommand{\Zetait}{\unicode{x1D6E7}}\)
\(\newcommand{\Etait}{\unicode{x1D6E8}}\)
\(\newcommand{\Thetait}{\unicode{x1D6E9}}\)
\(\newcommand{\Varthetait}{\unicode{x1D6F3}}\)
\(\newcommand{\Iotait}{\unicode{x1D6EA}}\)
\(\newcommand{\Kappait}{\unicode{x1D6EB}}\)
\(\newcommand{\Lambdait}{\unicode{x1D6EC}}\)
\(\newcommand{\Muit}{\unicode{x1D6ED}}\)
\(\newcommand{\Nuit}{\unicode{x1D6EE}}\)
\(\newcommand{\Xiit}{\unicode{x1D6EF}}\)
\(\newcommand{\Omicronit}{\unicode{x1D6F0}}\)
\(\newcommand{\Piit}{\unicode{x1D6F1}}\)
\(\newcommand{\Rhoit}{\unicode{x1D6F2}}\)
\(\newcommand{\Sigmait}{\unicode{x1D6F4}}\)
\(\newcommand{\Tauit}{\unicode{x1D6F5}}\)
\(\newcommand{\Upsilonit}{\unicode{x1D6F6}}\)
\(\newcommand{\Phiit}{\unicode{x1D6F7}}\)
\(\newcommand{\Chiit}{\unicode{x1D6F8}}\)
\(\newcommand{\Psiit}{\unicode{x1D6F9}}\)
\(\newcommand{\Omegait}{\unicode{x1D6FA}}\)
\(\let \digammaup \Digammaup \)
\(\renewcommand {\digammait }{\mathit {\digammaup }}\)
\(\newcommand {\smallin }{\unicode {x220A}}\)
\(\newcommand {\smallowns }{\unicode {x220D}}\)
\(\newcommand {\notsmallin }{\LWRoverlaysymbols {/}{\unicode {x220A}}}\)
\(\newcommand {\notsmallowns }{\LWRoverlaysymbols {/}{\unicode {x220D}}}\)
\(\newcommand {\rightangle }{\unicode {x221F}}\)
\(\newcommand {\intclockwise }{\unicode {x2231}}\)
\(\newcommand {\ointclockwise }{\unicode {x2232}}\)
\(\newcommand {\ointctrclockwise }{\unicode {x2233}}\)
\(\newcommand {\oiint }{\unicode {x222F}}\)
\(\newcommand {\oiiint }{\unicode {x2230}}\)
\(\newcommand {\ddag }{\unicode {x2021}}\)
\(\newcommand {\P }{\unicode {x00B6}}\)
\(\newcommand {\copyright }{\unicode {x00A9}}\)
\(\newcommand {\dag }{\unicode {x2020}}\)
\(\newcommand {\pounds }{\unicode {x00A3}}\)
\(\newcommand {\iddots }{\unicode {x22F0}}\)
\(\newcommand {\utimes }{\overline {\times }}\)
\(\newcommand {\dtimes }{\underline {\times }}\)
\(\newcommand {\udtimes }{\overline {\underline {\times }}}\)
\(\newcommand {\leftwave }{\left \{}\)
\(\newcommand {\rightwave }{\right \}}\)
\(\newcommand {\toprule }[1][]{\hline }\)
\(\let \midrule \toprule \)
\(\let \bottomrule \toprule \)
\(\newcommand {\cmidrule }[2][]{}\)
\(\newcommand {\morecmidrules }{}\)
\(\newcommand {\specialrule }[3]{\hline }\)
\(\newcommand {\addlinespace }[1][]{}\)
\(\newcommand {\LWRsubmultirow }[2][]{#2}\)
\(\newcommand {\LWRmultirow }[2][]{\LWRsubmultirow }\)
\(\newcommand {\multirow }[2][]{\LWRmultirow }\)
\(\newcommand {\mrowcell }{}\)
\(\newcommand {\mcolrowcell }{}\)
\(\newcommand {\STneed }[1]{}\)
\( \newcommand {\multicolumn }[3]{#3}\)
\(\newcommand {\tothe }[1]{^{#1}}\)
\(\newcommand {\raiseto }[2]{{#2}^{#1}}\)
\(\newcommand {\ang }[2][]{(\mathrm {#2})\degree }\)
\(\newcommand {\num }[2][]{\mathrm {#2}}\)
\(\newcommand {\si }[2][]{\mathrm {#2}}\)
\(\newcommand {\LWRSI }[2][]{\mathrm {#1\LWRSInumber \,#2}}\)
\(\newcommand {\SI }[2][]{\def \LWRSInumber {#2}\LWRSI }\)
\(\newcommand {\numlist }[2][]{\mathrm {#2}}\)
\(\newcommand {\numrange }[3][]{\mathrm {#2\,\unicode {x2013}\,#3}}\)
\(\newcommand {\SIlist }[3][]{\mathrm {#2\,#3}}\)
\(\newcommand {\SIrange }[4][]{\mathrm {#2\,#4\,\unicode {x2013}\,#3\,#4}}\)
\(\newcommand {\tablenum }[2][]{\mathrm {#2}}\)
\(\newcommand {\ampere }{\mathrm {A}}\)
\(\newcommand {\candela }{\mathrm {cd}}\)
\(\newcommand {\kelvin }{\mathrm {K}}\)
\(\newcommand {\kilogram }{\mathrm {kg}}\)
\(\newcommand {\metre }{\mathrm {m}}\)
\(\newcommand {\mole }{\mathrm {mol}}\)
\(\newcommand {\second }{\mathrm {s}}\)
\(\newcommand {\becquerel }{\mathrm {Bq}}\)
\(\newcommand {\degreeCelsius }{\unicode {x2103}}\)
\(\newcommand {\coulomb }{\mathrm {C}}\)
\(\newcommand {\farad }{\mathrm {F}}\)
\(\newcommand {\gray }{\mathrm {Gy}}\)
\(\newcommand {\hertz }{\mathrm {Hz}}\)
\(\newcommand {\henry }{\mathrm {H}}\)
\(\newcommand {\joule }{\mathrm {J}}\)
\(\newcommand {\katal }{\mathrm {kat}}\)
\(\newcommand {\lumen }{\mathrm {lm}}\)
\(\newcommand {\lux }{\mathrm {lx}}\)
\(\newcommand {\newton }{\mathrm {N}}\)
\(\newcommand {\ohm }{\mathrm {\Omega }}\)
\(\newcommand {\pascal }{\mathrm {Pa}}\)
\(\newcommand {\radian }{\mathrm {rad}}\)
\(\newcommand {\siemens }{\mathrm {S}}\)
\(\newcommand {\sievert }{\mathrm {Sv}}\)
\(\newcommand {\steradian }{\mathrm {sr}}\)
\(\newcommand {\tesla }{\mathrm {T}}\)
\(\newcommand {\volt }{\mathrm {V}}\)
\(\newcommand {\watt }{\mathrm {W}}\)
\(\newcommand {\weber }{\mathrm {Wb}}\)
\(\newcommand {\day }{\mathrm {d}}\)
\(\newcommand {\degree }{\mathrm {^\circ }}\)
\(\newcommand {\hectare }{\mathrm {ha}}\)
\(\newcommand {\hour }{\mathrm {h}}\)
\(\newcommand {\litre }{\mathrm {l}}\)
\(\newcommand {\liter }{\mathrm {L}}\)
\(\newcommand {\arcminute }{^\prime }\)
\(\newcommand {\minute }{\mathrm {min}}\)
\(\newcommand {\arcsecond }{^{\prime \prime }}\)
\(\newcommand {\tonne }{\mathrm {t}}\)
\(\newcommand {\astronomicalunit }{au}\)
\(\newcommand {\atomicmassunit }{u}\)
\(\newcommand {\bohr }{\mathit {a}_0}\)
\(\newcommand {\clight }{\mathit {c}_0}\)
\(\newcommand {\dalton }{\mathrm {D}_\mathrm {a}}\)
\(\newcommand {\electronmass }{\mathit {m}_{\mathrm {e}}}\)
\(\newcommand {\electronvolt }{\mathrm {eV}}\)
\(\newcommand {\elementarycharge }{\mathit {e}}\)
\(\newcommand {\hartree }{\mathit {E}_{\mathrm {h}}}\)
\(\newcommand {\planckbar }{\mathit {\unicode {x210F}}}\)
\(\newcommand {\angstrom }{\mathrm {\unicode {x212B}}}\)
\(\let \LWRorigbar \bar \)
\(\newcommand {\bar }{\mathrm {bar}}\)
\(\newcommand {\barn }{\mathrm {b}}\)
\(\newcommand {\bel }{\mathrm {B}}\)
\(\newcommand {\decibel }{\mathrm {dB}}\)
\(\newcommand {\knot }{\mathrm {kn}}\)
\(\newcommand {\mmHg }{\mathrm {mmHg}}\)
\(\newcommand {\nauticalmile }{\mathrm {M}}\)
\(\newcommand {\neper }{\mathrm {Np}}\)
\(\newcommand {\yocto }{\mathrm {y}}\)
\(\newcommand {\zepto }{\mathrm {z}}\)
\(\newcommand {\atto }{\mathrm {a}}\)
\(\newcommand {\femto }{\mathrm {f}}\)
\(\newcommand {\pico }{\mathrm {p}}\)
\(\newcommand {\nano }{\mathrm {n}}\)
\(\newcommand {\micro }{\mathrm {\unicode {x00B5}}}\)
\(\newcommand {\milli }{\mathrm {m}}\)
\(\newcommand {\centi }{\mathrm {c}}\)
\(\newcommand {\deci }{\mathrm {d}}\)
\(\newcommand {\deca }{\mathrm {da}}\)
\(\newcommand {\hecto }{\mathrm {h}}\)
\(\newcommand {\kilo }{\mathrm {k}}\)
\(\newcommand {\mega }{\mathrm {M}}\)
\(\newcommand {\giga }{\mathrm {G}}\)
\(\newcommand {\tera }{\mathrm {T}}\)
\(\newcommand {\peta }{\mathrm {P}}\)
\(\newcommand {\exa }{\mathrm {E}}\)
\(\newcommand {\zetta }{\mathrm {Z}}\)
\(\newcommand {\yotta }{\mathrm {Y}}\)
\(\newcommand {\percent }{\mathrm {\%}}\)
\(\newcommand {\meter }{\mathrm {m}}\)
\(\newcommand {\metre }{\mathrm {m}}\)
\(\newcommand {\gram }{\mathrm {g}}\)
\(\newcommand {\kg }{\kilo \gram }\)
\(\newcommand {\of }[1]{_{\mathrm {#1}}}\)
\(\newcommand {\squared }{^2}\)
\(\newcommand {\square }[1]{\mathrm {#1}^2}\)
\(\newcommand {\cubed }{^3}\)
\(\newcommand {\cubic }[1]{\mathrm {#1}^3}\)
\(\newcommand {\per }{/}\)
\(\newcommand {\celsius }{\unicode {x2103}}\)
\(\newcommand {\fg }{\femto \gram }\)
\(\newcommand {\pg }{\pico \gram }\)
\(\newcommand {\ng }{\nano \gram }\)
\(\newcommand {\ug }{\micro \gram }\)
\(\newcommand {\mg }{\milli \gram }\)
\(\newcommand {\g }{\gram }\)
\(\newcommand {\kg }{\kilo \gram }\)
\(\newcommand {\amu }{\mathrm {u}}\)
\(\newcommand {\nm }{\nano \metre }\)
\(\newcommand {\um }{\micro \metre }\)
\(\newcommand {\mm }{\milli \metre }\)
\(\newcommand {\cm }{\centi \metre }\)
\(\newcommand {\dm }{\deci \metre }\)
\(\newcommand {\m }{\metre }\)
\(\newcommand {\km }{\kilo \metre }\)
\(\newcommand {\as }{\atto \second }\)
\(\newcommand {\fs }{\femto \second }\)
\(\newcommand {\ps }{\pico \second }\)
\(\newcommand {\ns }{\nano \second }\)
\(\newcommand {\us }{\micro \second }\)
\(\newcommand {\ms }{\milli \second }\)
\(\newcommand {\s }{\second }\)
\(\newcommand {\fmol }{\femto \mol }\)
\(\newcommand {\pmol }{\pico \mol }\)
\(\newcommand {\nmol }{\nano \mol }\)
\(\newcommand {\umol }{\micro \mol }\)
\(\newcommand {\mmol }{\milli \mol }\)
\(\newcommand {\mol }{\mol }\)
\(\newcommand {\kmol }{\kilo \mol }\)
\(\newcommand {\pA }{\pico \ampere }\)
\(\newcommand {\nA }{\nano \ampere }\)
\(\newcommand {\uA }{\micro \ampere }\)
\(\newcommand {\mA }{\milli \ampere }\)
\(\newcommand {\A }{\ampere }\)
\(\newcommand {\kA }{\kilo \ampere }\)
\(\newcommand {\ul }{\micro \litre }\)
\(\newcommand {\ml }{\milli \litre }\)
\(\newcommand {\l }{\litre }\)
\(\newcommand {\hl }{\hecto \litre }\)
\(\newcommand {\uL }{\micro \liter }\)
\(\newcommand {\mL }{\milli \liter }\)
\(\newcommand {\L }{\liter }\)
\(\newcommand {\hL }{\hecto \liter }\)
\(\newcommand {\mHz }{\milli \hertz }\)
\(\newcommand {\Hz }{\hertz }\)
\(\newcommand {\kHz }{\kilo \hertz }\)
\(\newcommand {\MHz }{\mega \hertz }\)
\(\newcommand {\GHz }{\giga \hertz }\)
\(\newcommand {\THz }{\tera \hertz }\)
\(\newcommand {\mN }{\milli \newton }\)
\(\newcommand {\N }{\newton }\)
\(\newcommand {\kN }{\kilo \newton }\)
\(\newcommand {\MN }{\mega \newton }\)
\(\newcommand {\Pa }{\pascal }\)
\(\newcommand {\kPa }{\kilo \pascal }\)
\(\newcommand {\MPa }{\mega \pascal }\)
\(\newcommand {\GPa }{\giga \pascal }\)
\(\newcommand {\mohm }{\milli \ohm }\)
\(\newcommand {\kohm }{\kilo \ohm }\)
\(\newcommand {\Mohm }{\mega \ohm }\)
\(\newcommand {\pV }{\pico \volt }\)
\(\newcommand {\nV }{\nano \volt }\)
\(\newcommand {\uV }{\micro \volt }\)
\(\newcommand {\mV }{\milli \volt }\)
\(\newcommand {\V }{\volt }\)
\(\newcommand {\kV }{\kilo \volt }\)
\(\newcommand {\W }{\watt }\)
\(\newcommand {\uW }{\micro \watt }\)
\(\newcommand {\mW }{\milli \watt }\)
\(\newcommand {\kW }{\kilo \watt }\)
\(\newcommand {\MW }{\mega \watt }\)
\(\newcommand {\GW }{\giga \watt }\)
\(\newcommand {\J }{\joule }\)
\(\newcommand {\uJ }{\micro \joule }\)
\(\newcommand {\mJ }{\milli \joule }\)
\(\newcommand {\kJ }{\kilo \joule }\)
\(\newcommand {\eV }{\electronvolt }\)
\(\newcommand {\meV }{\milli \electronvolt }\)
\(\newcommand {\keV }{\kilo \electronvolt }\)
\(\newcommand {\MeV }{\mega \electronvolt }\)
\(\newcommand {\GeV }{\giga \electronvolt }\)
\(\newcommand {\TeV }{\tera \electronvolt }\)
\(\newcommand {\kWh }{\kilo \watt \hour }\)
\(\newcommand {\F }{\farad }\)
\(\newcommand {\fF }{\femto \farad }\)
\(\newcommand {\pF }{\pico \farad }\)
\(\newcommand {\K }{\mathrm {K}}\)
\(\newcommand {\dB }{\mathrm {dB}}\)
\(\newcommand {\kibi }{\mathrm {Ki}}\)
\(\newcommand {\mebi }{\mathrm {Mi}}\)
\(\newcommand {\gibi }{\mathrm {Gi}}\)
\(\newcommand {\tebi }{\mathrm {Ti}}\)
\(\newcommand {\pebi }{\mathrm {Pi}}\)
\(\newcommand {\exbi }{\mathrm {Ei}}\)
\(\newcommand {\zebi }{\mathrm {Zi}}\)
\(\newcommand {\yobi }{\mathrm {Yi}}\)
\(\require {mhchem}\)
\(\require {cancel}\)
\(\newcommand {\fint }{âĺŊ}\)
\(\newcommand {\hdots }{\cdots }\)
\(\newcommand {\mathnormal }[1]{#1}\)
\(\newcommand {\vecs }[2]{\vec {#1}_{#2}}\)
\(\newcommand {\KK }{\mathbb {K}}\)
\(\renewcommand {\C }{\mathcal {C}}\)
\(\newcommand {\D }{\mathcal {D}}\)
\(\renewcommand {\F }{\mathcal {F}}\)
\(\renewcommand {\P }{\mathcal {P}}\)
\(\newcommand {\R }{\mathcal {R}}\)
\(\renewcommand {\S }{\mathcal {S}}\)
\(\newcommand {\orth }{\mathbin {\bot }}\)
\(\newcommand {\iu }{\mathrm {i}}\)
\(\newcommand {\unif }{\text {unif}}\)
\(\newcommand {\supp }{\operatorname {supp}}\)
\(\newcommand {\Kern }{\text {Kern}}\)
\(\newcommand {\Bild }{\text {Bild}}\)
\(\newcommand {\Lin }{\operatorname {Lin}}\)
\(\newcommand {\id }{\text {id}}\)
\(\newcommand {\dist }{\operatorname {dist}}\)
\(\renewcommand {\Re }{\operatorname {Re}}\)
\(\renewcommand {\Im }{\operatorname {Im}}\)
\(\renewcommand {\graph }{\text {graph}}\)
\(\renewcommand {\div }{\operatorname {div}}\)
\(\newcommand {\loc }{{\text {loc}}}\)
\(\newcommand {\name }[1]{\textsc {#1}}\)
\(\newcommand {\smallpmatrix }[1]{\left (\begin {smallmatrix}#1\end {smallmatrix}\right )}\)
\(\newcommand {\matlab }{{\fontfamily {bch}\scshape \selectfont {}Matlab}}\)
\(\newcommand {\innerproduct }[1]{\left \langle {#1}\right \rangle }\)
\(\newcommand {\norm }[1]{\left \Vert {#1}\right \Vert }\)
\(\renewcommand {\natural }{\mathbb {N}}\)
\(\newcommand {\integer }{\mathbb {Z}}\)
\(\newcommand {\rational }{\mathbb {Q}}\)
\(\newcommand {\real }{\mathbb {R}}\)
\(\newcommand {\complex }{\mathbb {C}}\)
\(\renewcommand {\d }{\mathop {}\!\mathrm {d}}\)
\(\newcommand {\dr }{\d {}r}\)
\(\newcommand {\ds }{\d {}s}\)
\(\newcommand {\dt }{\d {}t}\)
\(\newcommand {\du }{\d {}u}\)
\(\newcommand {\dv }{\d {}v}\)
\(\newcommand {\dw }{\d {}w}\)
\(\newcommand {\dx }{\d {}x}\)
\(\newcommand {\dy }{\d {}y}\)
\(\newcommand {\dz }{\d {}z}\)
\(\newcommand {\dsigma }{\d {}\sigma }\)
\(\newcommand {\dphi }{\d {}\phi }\)
\(\newcommand {\dvarphi }{\d {}\varphi }\)
\(\newcommand {\dtau }{\d {}\tau }\)
\(\newcommand {\dxi }{\d {}\xi }\)
\(\newcommand {\dtheta }{\d {}\theta }\)
\(\newcommand {\tp }{\mathrm {T}}\)
Wiederholung und Motivation
Bemerkung: Zur Wiederholung werden die Definitionen von Sobolev- und Hölderräumen wiedergegeben.
Sobolevraum: Seien \(\Omega \subset \real ^n\) offen, \(m \in \natural _0\) und \(p \in [1, \infty ]\).
Dann heißt der Vektorraum \(W^{m,p}(\Omega ) := \{f \in L^p(\Omega ) \;|\; \forall _{s \in \natural _0^n,\, |s| \le m} \exists _{f^{(s)} \in L^p(\Omega )}\; f^{(0)} = f,\)
\(\forall _{\varphi \in \C ^\infty _c(\Omega )}\; \int _\Omega (\partial _x^s \varphi ) f \dx = (-1)^{|s|} \int _\Omega \varphi f^{(s)} \dx \}\) Sobolevraum der Ordnung \(m\) mit Exponent \(p\).
\(W^{m,p}(\Omega )\) wird mit der Norm \(\norm {f}_{W^{m,p}(\Omega )} := \sum _{|s| \le m} \norm {f^{(s)}}_{L^p(\Omega )}\) versehen. Für \(p = 2\) schreibt man auch \(H^m(\Omega ) :=
W^{m,2}(\Omega )\) bzw. \(\norm {\cdot }_{H^m(\Omega )} := \norm {\cdot }_{W^{m,2}(\Omega )}\). Die Funktionen \(f^{(s)}\) für \(|s| \ge 1\) heißen schwache
Ableitungen von \(f\) und werden mit \(\partial _x^s f := f^{(s)}\) bezeichnet.
Bemerkung: Es gilt \(W^{m,p}(\Omega ) = \overline {W^{m,p}(\Omega ) \cap \C ^\infty (\Omega )}^{\norm {\cdot }_{W^{m,p}(\Omega )}}\) für \(p < \infty \).
\(W^{m,p}_0(\Omega ) := \overline {\C ^\infty _c(\Omega )}^{\norm {\cdot }_{W^{m,p}(\Omega )}}\) ist der Sobolevraum mit (verallg.)
Nullrandwerten.
Hölderraum: Seien \(\Omega \subset \real ^n\) offen oder kompakt, \(k \in \natural _0\) und \(\alpha \in (0, 1]\).
\(\C ^{k,\alpha }(\Omega ) := \{f \in \C ^k_b(\Omega ) \;|\; \partial _x^j f \in \C ^{0,\alpha }(\Omega ) \text { für } |j| = k\}\) heißt Hölderraum der Ordnung \(k\) mit Exponent \(\alpha \). \(\C ^{k,\alpha }(\Omega )\) wird mit der Norm \(\norm {f}_{\C ^{k,\alpha }(\Omega )} := \norm {f}_{\C ^k(\Omega )} +
\sum _{|j|=k} [\partial _x^j f]_{\C ^{0,\alpha }(\Omega )}\) versehen, wobei \([f]_{\C ^{0,\alpha }(\Omega )} := \sup _{x_1, x_2 \in \Omega ,\; x_1 \not = x_2} \frac {|f(x_1) - f(x_2)|}{|x_1 -
x_2|^\alpha }\). (Für \(\alpha = 0\) definiert man \(\C ^{k,0}(\Omega ) := \C ^k(\Omega )\).)
Bemerkung: Gesucht sind Bedingungen an \(n, m, p, k, \alpha \), sodass \(W^{m,p}(\real ^n) \subset \C ^{k,\alpha }(\real ^n)\)
(oder sodass \(W^{m,p}(\Omega ) \subset \C ^{k,\alpha }(\overline {\Omega })\) mit \(\Omega \subset \real ^n\) offen, beschränkt, Lipschitz-berandet).
Für \(u \in W^{1,\infty }(\real ^n)\) gilt \(\sup _{x_1, x_2 \in \real ^n,\; x_1 \not = x_2} \frac {|u(x_1) - u(x_2)|}{|x_1 - x_2|} \le \sup _{x \in \real ^n} |\nabla u(x)|\) nach dem
Mittelwertsatz (auch Hauptsatz der Differentialrechnung). Für \(u \in W^{1,\infty }(\real ^n)\) gilt also \(u \in \C ^{0,1}(\real ^n)\), d. h. für den Fall \((m, p, k, \alpha ) = (1,
\infty , 0, 1)\) gilt \(W^{m,p}(\real ^n) \subset \C ^{k,\alpha }(\real ^n)\).
Man kann zeigen: Allgemeiner existieren für bestimmte \(\alpha \in (0, 1)\) und \(p \in [1, \infty )\) auch Ungleichungen der Form \([u]_{\C ^{0,\alpha }(\real ^n)} \le C(n,p) \norm {\nabla
u}_{L^p(\real ^n)}\) (Fall \(m = 1\), \(k = 0\)).
Um an die Beziehung zwischen \(\alpha , n, p\) zu gelangen, bedient man sich eines Skalierungsarguments. Angenommen, eine solche Ungleichung existiert für \(u \in W^{1,p}(\real ^n)\). Dann ist für
\(\lambda > 0\) auch \(u_\lambda \in W^{1,p}(\real ^n)\) mit \(u_\lambda (x) := u(\frac {x}{\lambda })\). Es gilt \([u_\lambda ]_{\C ^{0,\alpha }(\real ^n)} = \sup _{x_1 \not = x_2} \frac
{|u_\lambda (x_1) - u_\lambda (x_2)|}{|x_1 - x_2|^\alpha }\)
\(= \lambda ^{-\alpha } \cdot \sup _{x_1 \not = x_2} \frac {|u(x_1/\lambda ) - u(x_2/\lambda )|}{|x_1/\lambda - x_2/\lambda |^\alpha } = \lambda ^{-\alpha } \cdot [u]_{\C ^{0,\alpha }(\real
^n)}\) sowie
\(\norm {\nabla u_\lambda }_{L^p(\real ^n)} = \left (\int _{\real ^n} |\nabla u_\lambda (x)|^p \dx \right )^{1/p} = \left (\int _{\real ^n} \lambda ^{-p} |\nabla u(\frac {x}{\lambda })|^p \dx
\right )^{1/p} = \left (\int _{\real ^n} \lambda ^{n-p} |\nabla u(y)|^p \dy \right )^{1/p}\)
\(= \lambda ^{n/p-1} \cdot \norm {\nabla u}_{L^p(\real ^n)}\). Unter der Annahme der Existenz der obigen Ungleichung gilt damit
\(\lambda ^{-\alpha } \cdot [u]_{\C ^{0,\alpha }(\real ^n)} = [u_\lambda ]_{\C ^{0,\alpha }(\real ^n)} \le C(n,p) \cdot \norm {\nabla u_\lambda }_{L^p(\real ^n)} = C(n,p) \lambda ^{n/p-1}
\cdot \norm {\nabla u}_{L^p(\real ^n)}\) bzw.
\([u]_{\C ^{0,\alpha }(\real ^n)} \le \lambda ^{n/p-1+\alpha } \cdot C(n,p) \norm {\nabla u}_{L^p(\real ^n)}\). Diese Ungleichung kann nur für alle \(\lambda > 0\) gelten, wenn \(\frac
{n}{p} - 1 + \alpha = 0\) ist, also \(1 - \frac {n}{p} = \alpha \). Insbesondere muss wegen \(\alpha > 0\) auch \(p > n\) gelten.
Für höhere Ableitungen (\(m > 1\) oder \(k > 0\)) verfährt man ähnlich.
Man vermutet daher, dass \(W^{m,p}(\real ^n) \subset \C ^{k,\alpha }(\real ^n)\) für \(m \in \natural \), \(p \in [1, \infty )\), \(k \in \natural _0\) und \(\alpha \in (0, 1)\) mit \(m - \frac
{n}{p} = k + \alpha \).
Bemerkung: Gesucht sind Bedingungen an \(n, m_1, p_1, m_2, p_2\), sodass \(W^{m_1,p_1}(\real ^n) \subset W^{m_2,p_2}(\real ^n)\) (oder sodass \(W^{m_1,p_1}(\Omega ) \subset W^{m_2,p_2}(\Omega
)\) mit \(\Omega \subset \real ^n\) offen, beschränkt, Lipschitz-berandet).
Für \(1 \le p < n\) kann man zeigen, dass es ein \(p^\ast > p\) gibt mit \(\norm {u}_{L^{p^\ast }(\real ^n)} \le C(n,p) \norm {\nabla u}_{L^p(\real ^n)}\) für alle \(u \in
W^{1,p}(\real ^n)\). Daraus folgt dann direkt \(W^{1,p}(\real ^n) \subset L^{p^\ast }(\real ^n)\) (Fall \(m_1 = 1\), \(m_2 = 0\)).
Zur Bestimmung von \(p^\ast \) benutzt man wieder obiges Reskalierungsargument:
\(\norm {u_\lambda }_{L^{p^\ast }(\real ^n)} = \lambda ^{n/p^\ast } \cdot \norm {u}_{L^{p^\ast }(\real ^n)}\) und \(\norm {\nabla u_\lambda }_{L^p(\real ^n)} = \lambda ^{n/p-1} \cdot \norm
{\nabla u}_{L^p(\real ^n)}\) wie oben.
Damit gilt \(\lambda ^{n/p^\ast } \cdot \norm {u}_{L^{p^\ast }(\real ^n)} = \norm {u_\lambda }_{L^{p^\ast }(\real ^n)} \le C(n,p) \norm {\nabla u_\lambda }_{L^p(\real ^n)} = \lambda ^{n/p-1}
\cdot C(n,p) \norm {\nabla u}_{L^p(\real ^n)}\), also \(\norm {u}_{L^{p^\ast }(\real ^n)} \le \lambda ^{n/p-1-n/p^\ast } \cdot C(n,p) \norm {\nabla u}_{L^p(\real ^n)}\) für alle \(\lambda
> 0\). Daraus folgt \(\frac {n}{p} - 1 - \frac {n}{p^\ast } = 0\) bzw. \(1 - \frac {n}{p} = -\frac {n}{p^\ast } \iff p^\ast = \frac {np}{n - p}\).
Als Verallgemeinerung vermutet man \(W^{m_1,p_1}(\real ^n) \subset W^{m_2,p_2}(\real ^n)\) für bestimmte \(m_1, m_2 \in \natural _0\) mit \(m_1 \ge m_2\) und \(p_1, p_2 \in [1, \infty )\)
(genauer: für \(m_1 - \frac {n}{p_1} = m_2 - \frac {n}{p_2}\) und \(m_1 \ge m_2\)).
Beispiel: Wie hoch muss \(m \in \natural \) sein, damit \(H^m(\real ^3) \subset \C ^2(\real ^3)\)? (Zunächst sollen nur die Einbettungen \(W^{1,p}(\real ^n) \subset \C ^{0,\alpha }(\real
^n)\) und \(W^{1,p}(\real ^n) \subset L^{p^\ast }(\real ^n)\) benutzt werden.)
Sei \(u \in H^m(\real ^3)\). Dann existieren die schwachen Ableitungen \(\partial _x^j u \in L^2(\real ^3)\) in den Ordnungen \(|j| \le m\). Für \(H^m(\real ^3)\) ist \(p = 2\) und damit kleiner als \(n =
3\). Daher kann die erste Einbettung aus den Bemerkungen oben nicht verwendet werden. Stattdessen kann man die zweite Einbettung \(W^{1,p}(\real ^n) \subset L^{p^\ast }(\real ^n)\) verwenden. Es gilt \(p^\ast =
\frac {np}{n - p} = \frac {3 \cdot 2}{3 - 2} = 6\), also \(H^1(\real ^3) \subset L^6(\real ^3)\). Wegen \(\forall _{|j| \le m-1}\; \partial _x^j u \in H^1(\real ^3)\) gilt daher \(\partial _x^j u \in
L^6(\real ^3)\) für alle \(|j| \le m - 1\), also \(u \in W^{m-1,6}(\real ^3)\).
Nun gilt \(p^\ast > n\), daher kann man jetzt die erste Einbettung verwenden (für \(m’ := m - 1\)). Aus der Gleichung \((m-1) - \frac {n}{p^\ast } = k + \alpha \) errechnet man \(\alpha = (m-1) -
\frac {n}{p^\ast } - k = (m-1) - \frac {3}{6} - 2 \in (0, 1)\) zum Beispiel für \((m-1) = 3\) (mit dem gewünschten \(k = 2\)). Damit gilt \(W^{3,6}(\real ^3) \subset \C ^{2,1/2}(\real
^3)\).
Insgesamt gilt also \(H^m(\real ^3) \subset H^4(\real ^3) \subset W^{3,6}(\real ^3) \subset \C ^{2,1/2}(\real ^3) \subset \C ^2(\real ^3)\) für \(m \ge 4\).
Wenn man \(W^{m,p}(\real ^n) \subset \C ^{k,\alpha }(\real ^n)\) mit \(m - \frac {n}{p} = k + \alpha \) verwendet, so erhält man das Resultat direkt (mit \((n, m, p, k, \alpha ) = (3, 4, 2, 2,
\frac {1}{2})\)).
Bemerkung: Was kann man für beschränkte Gebiete erwarten?
Sei \(f_\varrho \colon \overline {B_1(0)} \subset \real ^n \to \real \) mit \(f_\varrho (x) := |x|^\varrho \) für \(x \not = 0\) und \(f_\varrho (0) := 0\), wobei \(\varrho \in \real
\setminus \natural _0\).
Man kann direkt nachrechnen, dass dann gilt:
Für \(k \in \natural _0\) und \(\alpha \in (0, 1]\) gilt \(f_\varrho \in \C ^{k,\alpha }(\overline {B_1(0)}) \iff \varrho \ge k + \alpha \).
Für \(m \in \natural _0\) und \(p \in [1, \infty )\) gilt \(f_\varrho \in W^{m,p}(B_1(0)) \iff \varrho \ge m - \frac {n}{p}\).
Dies motiviert die Vermutungen
\(W^{m_1,p_1}(B_1(0)) \subset W^{m_2,p_2}(B_1(0))\) für \(m_1 - \frac {n}{p_1} \ge m_2 - \frac {n}{p_2}\), \(m_1 \ge m_2\) und \(p_1, p_2 \in [1, \infty )\) sowie
\(W^{m,p}(B_1(0)) \subset \C ^{k,\alpha }(\overline {B_1(0)})\) für \(m - \frac {n}{p} \ge k + \alpha \), \(p \in [1, \infty )\) und \(\alpha \in (0, 1)\).
Gagliardo-Nirenberg-Sobolev-Ungleichung
Bemerkung: Die Gagliardo-Nirenberg-Sobolev-Ungleichung beweist durch das anschließende Korollar die Einbettung \(W^{m_1,p_1}(\real ^n) \subset W^{m_2,p_2}(\real ^n)\) für den Fall \(m_1 = 1\),
\(m_2 = 0\).
Satz (Gagliardo-Nirenberg-Sobolev-Ungleichung):
Seien \(p \in [1, n)\), \(p^\ast := \frac {np}{n - p}\) und \(u \in \C ^1_c(\real ^n)\).
Dann ist \(u \in L^{p^\ast }(\real ^n)\) mit \(\norm {u}_{L^{p^\ast }(\real ^n)} \le C(n,p) \norm {\nabla u}_{L^p(\real ^n)}\).
Folgerung: Seien \(p \in [1, n)\), \(p^\ast := \frac {np}{n - p}\) und \(u \in W^{1,p}(\real ^n)\).
Dann ist \(u \in L^{p^\ast }(\real ^n)\) mit \(\norm {u}_{L^{p^\ast }(\real ^n)} \le C(n,p) \norm {\nabla u}_{L^p(\real ^n)}\).
Bemerkung: Für den Beweis des Korollars muss man Glättung durch Faltung (wenn \(u\) kompakten Träger besitzt) und Abschneiden durch Multiplikation (wenn \(u\) keinen kompakten
Träger besitzt) durchführen.
Lemma (Approximation durch Faltung): Seien \(\varphi \in \C ^\infty _c(\real ^n)\) mit
\(\forall _{y \in \real ^n}\; \varphi (y) \ge 0,\; \varphi (-y) = \varphi (y)\) und \(\int _{\real ^n} \varphi (y)\dy = 1\) sowie \(\varphi _\varepsilon (x) := \varepsilon ^{-n} \varphi (\frac
{x}{\varepsilon })\) für \(\varepsilon > 0\).
Außerdem seien \(u \in L^p(\real ^n)\) und \(u_\varepsilon := \varphi _\varepsilon \ast u\). Dann gilt
\(\supp (\varphi \ast u) \subset \overline {\supp (\varphi ) + \supp (u)}\),
\(u_\varepsilon \in \C ^\infty (\real ^n)\) mit \(\partial _x^s u_\varepsilon = (\partial _x^s \varphi _\varepsilon ) \ast u\),
für \(u \in W^{1,p}(\real ^n)\) gilt \(\nabla u_\varepsilon = (\nabla u)_\varepsilon := \varphi _\varepsilon \ast \nabla u\),
-
(4)
\(\norm {u_\varepsilon }_{L^p(\real ^n)} \le \norm {u}_{L^p(\real ^n)}\) (wegen \(\norm {\varphi \ast u}_{L^p(\real ^n)} \le \norm {\varphi }_{L^1(\real ^n)} \norm {u}_{L^p(\real ^n)}\)) und
für \(u \in W^{1,p}(\real ^n)\) gilt \(\norm {(\nabla u)_\varepsilon }_{L^p(\real ^n)} \le \norm {\nabla u}_{L^p(\real ^n)}\)
und
-
(5)
\(\lim _{\varepsilon \to 0} \norm {u_\varepsilon - u}_{L^p(\real ^n)} = 0\),
damit gilt für \(u \in W^{1,p}(\real ^n)\), dass \(\lim _{\varepsilon \to 0} \norm {u_\varepsilon - u}_{W^{1,p}(\real ^n)} = 0\),
außerdem \(\forall _{R > 0}\; \lim _{\varepsilon \to 0} \norm {u_\varepsilon - u}_{L^1(B_R(0))} = 0\) und
damit \(u_\varepsilon \to u\) f.ü. in \(\real ^n\).
Lemma (Approximation durch Abschneidefunktionen):
Seien \(\eta \in \C ^\infty (\real ^n)\) mit \(\forall _{z \in \real ^n}\; \eta (z) \in [0, 1]\), \(\eta (z) = 1\) für alle \(|z| \le 1\) und \(\eta (z) = 0\) für alle \(|z| \ge 2\) sowie
\(\eta _R(z) := \eta (\frac {z}{R})\) für \(R > 0\). Außerdem seien \(u \in W^{1,p}(\real ^n)\) und \(u_R := \eta _R \cdot u\).
Dann gilt \(u_R \in W^{1,p}(\real ^n)\), wobei
\(\norm {u_R}_{L^p(\real ^n)} \le \norm {u}_{L^p(\real ^n)}\),
\(\norm {\nabla u_R}_{L^p(\real ^n)} \le \norm {\nabla u}_{L^p(\real ^n)} + \frac {1}{R} \norm {\nabla \eta }_{L^\infty (\real ^n)} \norm {u}_{L^p(\real ^n)}\) (wegen \(\nabla u_R =
\eta _R \nabla u + u \nabla \eta _R\)).
Teil 1 des Sobolevschen Einbettungssatzes
Lemma (Fortsetzungsoperator): Seien \(\Omega \subset \real ^n\) offen, beschränkt und Lipschitz-berandet,
\(p \in [1, \infty ]\) und \(\delta > 0\). Dann gibt es einen linearen und stetigen Fortsetzungsoperator
\(E\colon W^{1,p}(\Omega ) \to W_0^{1,p}(B_\delta (\Omega ))\) mit \(\forall _{u \in W^{1,p}(\Omega )}\; (Eu)|_\Omega = u\).
Satz (Teil 1 des Sobolevschen Einbettungssatzes):
Seien \(m_1, m_2 \in \natural _0\) und \(p_1, p_2 \in [1, \infty )\).
Ist \(m_1 - \frac {n}{p_1} = m_2 - \frac {n}{p_2}\) und \(m_1 \ge m_2\), dann existiert die Einbettung
\(\id \colon W^{m_1,p_1}(\real ^n) \to W^{m_2,p_2}(\real ^n)\) und ist stetig, d. h.
\(\exists _{C > 0} \forall _{u \in W^{m_1,p_1}(\real ^n)}\; \norm {u}_{W^{m_2,p_2}(\real ^n)} \le C \norm {u}_{W^{m_1,p_1}(\real ^n)}\) mit \(C = C(n, m_1, p_1, m_2, p_2)\).
Sei \(\Omega \subset \real ^n\) offen, beschränkt und Lipschitz-berandet.
Ist \(m_1 - \frac {n}{p_1} \ge m_2 - \frac {n}{p_2}\) und \(m_1 \ge m_2\), dann existiert die Einbettung
\(\id \colon W^{m_1,p_1}(\Omega ) \to W^{m_2,p_2}(\Omega )\) und ist stetig, d. h.
\(\exists _{C > 0} \forall _{u \in W^{m_1,p_1}(\Omega )}\; \norm {u}_{W^{m_2,p_2}(\Omega )} \le C \norm {u}_{W^{m_1,p_1}(\Omega )}\) mit \(C = C(\Omega , n, m_1, p_1, m_2, p_2)\).
Ist \(m_1 - \frac {n}{p_1} > m_2 - \frac {n}{p_2}\) und \(m_1 > m_2\), dann ist die Einbettung \(\id \colon W^{m_1,p_1}(\Omega ) \to W^{m_2,p_2}(\Omega )\) sogar kompakt.
Für \(\widetilde {\Omega } \subset \real ^n\) nur offen und beschränkt gelten die Aussagen (2) und (3) für die Räume
\(W^{m_i,p_i}_0(\widetilde {\Omega })\) anstatt \(W^{m_i,p_i}(\Omega )\), wobei \(W^{0,p}_0(\widetilde {\Omega }) := L^p(\widetilde {\Omega })\).
Morreysche Ungleichung
Bemerkung: Die Morreysche Ungleichung beweist durch den zweiten Teil des anschließenden Korollars die Einbettung \(W^{m,p}(\real ^n) \subset \C ^{k,\alpha }(\real ^n)\) für den Fall \(m = 1\),
\(k = 0\).
Satz (Morreysche Ungleichung): Seien \(p \in (n, \infty ]\), \(\alpha := 1 - \frac {n}{p}\) und \(u \in
\C ^1(\real ^n)\).
Dann ist \(u \in \C ^{0,\alpha }(\real ^n)\) mit \([u]_{\C ^{0,\alpha }(\real ^n)} \le C(n,p) \norm {\nabla u}_{L^p(\real ^n)}\).
Bemerkung: Die Bedingung \(p > n\) ist nötig, damit keine Singularitäten auftreten (sonst \(\alpha \le 0\)).
Hölder-stetig für \(L^p\)-Funktionen: Seien \(u \in L^p(\real ^n)\) und \(\alpha \in [0, 1]\).
Dann heißt \(u\) Hölder-stetig mit Exponent \(\alpha \) (\(u \in \C ^{0,\alpha }(\real ^n)\)), falls \(\exists _{\widetilde
{u} \in \C ^{0,\alpha }(\real ^n)}\; u = \widetilde {u} \text { f.ü. auf } \real ^n\). Außerdem sei \(\norm {u}_{\C ^{0,\alpha }(\real ^n)} := \norm {\widetilde {u}}_{\C ^{0,\alpha }(\real
^n)}\). Analog sind \(u \in \C ^{k,\alpha }(\real ^n)\) und \(\norm {u}_{\C ^{k,\alpha }(\real ^n)}\) für \(k \in \natural _0\) definiert. \(\real ^n\) kann durch \(\Omega \) für \(\Omega
\subset \real ^n\) offen ersetzt werden.
Folgerung: Seien \(p \in (n, \infty )\) und \(\alpha := 1 - \frac {n}{p}\).
Sei \(u \in L^1_\loc (\real ^n)\) mit \(\nabla u \in L^p(\real ^n)\).
Dann ist \(u \in \C ^{0,\alpha }(\real ^n)\) mit \([u]_{\C ^{0,\alpha }(\real ^n)} \le C(n,p) \norm {\nabla u}_{L^p(\real ^n)}\).
Sei \(u \in W^{1,p}(\real ^n)\).
Dann ist \(u \in \C ^{0,\alpha }(\real ^n)\) mit \(\norm {u}_{\C ^{0,\alpha }(\real ^n)} \le C(n,p) \norm {u}_{W^{1,p}(\real ^n)}\).
Teil 2 des Sobolevschen Einbettungssatzes
Lemma (Einbettungssätze für Hölder-Räume):
Sei \(\Omega \subset \real ^n\) offen, beschränkt und Lipschitz-berandet. Dann gilt:
Für \(k \in \natural _0\) ist die Einbettung \(\id \colon \C ^{k+1}(\overline {\Omega }) \to \C ^{k,1}(\overline {\Omega })\) stetig.
Seien \(k_1, k_2 \in \natural _0\) und \(\alpha _1, \alpha _2 \in [0, 1]\) mit \(k_1 + \alpha _1 > k_2 + \alpha _2\) (im Fall \(k_1 = 0\) kann sogar auf die Lipschitz-Berandung verzichtet
werden).
Dann ist die Einbettung \(\id \colon \C ^{k_1,\alpha _1}(\overline {\Omega }) \to \C ^{k_2,\alpha _2}(\overline {\Omega })\) kompakt, wobei \(\C ^{k,0}(\overline {\Omega }) := \C ^k(\overline
{\Omega })\).
Satz (Teil 2 des Sobolevschen Einbettungssatzes):
Seien \(m \in \natural \), \(p \in [1, \infty )\), \(k \in \natural _0\) und \(\alpha \in [0, 1]\).
Ist \(m - \frac {n}{p} = k + \alpha \) und \(\alpha \in (0, 1)\), dann existiert die Einbettung \(\id \colon W^{m,p}(\real ^n) \to \C ^{k,\alpha }(\real ^n)\) und ist stetig, d. h.
\(\exists _{C > 0} \forall _{u \in W^{m,p}(\real ^n)}\; \norm {u}_{\C ^{k,\alpha }(\real ^n)} \le C \norm {u}_{W^{m,p}(\real ^n)}\) mit \(C = C(n, m, p, k, \alpha )\).
Sei \(\Omega \subset \real ^n\) offen, beschränkt und Lipschitz-berandet.
Ist \(m - \frac {n}{p} \ge k + \alpha \) und \(\alpha \in (0, 1)\), dann existiert die Einbettung \(\id \colon W^{m,p}(\Omega ) \to \C ^{k,\alpha }(\overline {\Omega })\) und ist stetig, d. h.
\(\exists _{C > 0} \forall _{u \in W^{m,p}(\Omega )}\; \norm {u}_{\C ^{k,\alpha }(\overline {\Omega })} \le C \norm {u}_{W^{m,p}(\Omega )}\) mit \(C = C(\Omega , n, m, p, k, \alpha )\).
Ist \(m - \frac {n}{p} > k + \alpha \) und \(\alpha \in [0, 1]\), dann existiert die Einbettung \(\id \colon W^{m,p}(\Omega ) \to \C ^{k,\alpha }(\overline {\Omega })\) und ist stetig und
kompakt.
Für \(\widetilde {\Omega } \subset \real ^n\) nur offen und beschränkt gelten die Aussagen (2) und (3) für die Räume \(W^{m,p}_0(\widetilde
{\Omega })\) anstatt \(W^{m,p}(\Omega )\).
Satz (Einbettung für \(p = \infty \), \(\alpha = 1\) ist Isomorphismus):
Seien \(k \in \natural _0\) sowie \(\Omega \subset \real ^n\) offen, beschränkt und Lipschitz-berandet.
Dann ist die Einbettung \(\id \colon \C ^{k,1}(\overline {\Omega }) \to W^{k+1,\infty }(\Omega )\) ein Isomorphismus.