Introduction

Notes

Math

Epistemology

Search

Andrius Kulikauskas

  • m a t h 4 w i s d o m - g m a i l
  • +370 607 27 665
  • My work is in the Public Domain for all to share freely.

用中文

  • 读物 书 影片 维基百科

Introduction E9F5FC

Questions FFFFC0

Software


Bott periodicity, CT Groups

Understand the correspondence between the various manifestations of Bott periodicity


Bott Correspondence



Slides

  • 9+1, 8+2, 7+3, 6+4
  • two kinds of Clifford algebra
    • squaring products of Clifford generators
  • R,C,H structure
    • {$a^2=\theta$}, {$a^4=1$}
    • 2x2 matrices and the operators
    • three group extensions: recurring activity and structure
  • matrix expression of {$J_1, J_2, J_3\cdots$}

Readings

Questions

  • Understand real and quaternionic structures.
  • Understand the definition of {$(\phi, \chi )$}-representation.
  • Understand why complex linear operator CT is matched with the commutative group extension, and why complex linear operators C, T are matched with the noncommutative group extension.
  • Understand how to interpret C and iC as acting on the complex vector space.

Notation

{$ε±$} is odd and {$ε^2± = ±1$}

{$\eta=\mathbb{R}^{1|1}\;\;\;\rho(e)=\begin{pmatrix}0 & -1\\1 & 0 \\ \end{pmatrix} \;\;\;\; \tilde{\eta}=\mathbb{R}^{1|1}\;\;\;\rho(e)=\begin{pmatrix}0 & 1\\1 & 0 \\ \end{pmatrix}$}

{$\iota = \begin{pmatrix}0 & -1\\1 & 0 \\ \end{pmatrix} \;\;\;\; \varphi = \begin{pmatrix}1 & 0 \\ 0 & -1 \\ \end{pmatrix} \;\;\;\; \psi=\begin{pmatrix} 0 & 0 & 1 & 0 \\ 0 & 0 & 0 & -1 \\ -1 & 0 & 0 & 0 \\ 0 & 1 & 0 & 0 \\ \end{pmatrix}$}

Lie algebra decomposition {$\frak{g}=\frak{h}+\frak{m}$} where {$\frak{[h, h] ∈ h, [h, m] ∈ m, [m, m] ∈ h}$}.

{$m\in G/H$}

    symmetric space {$G/H$}Hamil tonian{$T^2$}{$C^2$}{$T$}{$C$}Restrict
{$iC$}{$C\ell_{+2}$}{${\small \iota\equiv j_1}$}{$C\ell_{0}$}{${\tiny O(16r)\times O(16r)}/O(16r)$}{$\iota\otimes m$} +1 {$\varphi\otimes \mathbb{I}$} 
{$iCT$}{$C\ell_{+3}$}{$J_1=$}{$\textrm{diag}(\iota)$}{$C\ell_{-1}$}{$O(16r)/U(8r)$}{$\iota\otimes m$}-1+1{$\varphi\otimes J_1$}{$\varphi$} 
{$CT$}?{$C\ell_4$}{$J_2=$}{$\textrm{diag}(\psi)$}{$C\ell_{-2}$}{$U(8r)/Sp(4r)$}{$J_1m$}-1 {$J_2$}  
{$iCT$}{$C\ell_{-3}$}{$J_3^{-1}=$}{$I_1J_1J_2$}{$C\ell_{-3}$}{$Sp(4r)/{\tiny Sp(2r)\times Sp(2r)}$}{$J_1m$}-1-1{$J_3$}{$J_2$} 
{$iC$}{$C\ell_{-2}$}{$J_4^{-1}=$}{$LJ_3$}{$C\ell_{-4}$}{${\tiny Sp(2r)\times Sp(2r)}/Sp(2r)$}{$J_1m$} -1 {$J_2$}{${\small J_1J_2J_3}$}{$=+1$}
{$C$} {$|_{T=+1}$}{$C\ell_{-1}$}{$J_5^{-1}=$}{$I_2J_1J_4$}{$C\ell_{-5}$}{$Sp(2r)/U(2r)$}{$J_1m$}+1-1{$J_2J_4J_5$}{$J_2$} 
{$|_{T=+1}$}{$C\ell_0$}{$J_6^{-1}=$}{$I_3J_2J_4$}{$C\ell_{-6}$}{$U(2r)/O(2r)$}{$J_1m$}+1 {${\small J_3J_4J_6}$} or {${\small J_2J_4J_6}$} {${\small J_1J_4J_5}$}{$=+1$}
{$C$} {$|_{T=+1}$}{$C\ell_{+1}$}{$J_7^{-1}=$}{$I_4J_1J_6$}{$C\ell_{-7}$}{$O(2r)/{\tiny O(r)\times O(r)}$}{$J_1m$}+1+1{$J_1J_6J_7$}{${\small J_2J_4J_6}$}{$\equiv\varphi $} 
{$iC$}{$C\ell_{+2}$}{$J_8^{-1}=$}{$L_2J_7$}{$C\ell_{-8}$}{${\tiny O(r)\times O(r)}/O(r)$}{$J_1m$} +1 {${\small J_2J_4J_6}$}{$\equiv\varphi$}{${\small J_2J_4J_6},$} {${\small J_1J_6J_7}$}{$=+1$}

We need to compare two Clifford algebras.

  • A Clifford algebra is generated by {$n$} mutually anticommuting linear complex structures, where {$G$} commutes with the first {$n-1$} linear complex structures and {$H$} commutes with all {$n$} linear complex structures. This pattern is centered on {$C$} and {$O(16r)$} and starts with {$O(16r)/U(8r)$}.
  • A Clifford algebra gets its generators from CT groups. This pattern is centered on {$T$} and {$U(2r)/O(2r)$}. The direction (forwards or backwards) is determined by {$C$} and the unfolding is determined by {$T$}.

{$T$} defines the {$J_k$} by acting as {$I_m$}, splitting the context into zones with eigenvalues {$\pm 1$}. {$C$} sets up an isometry {$L$}

It looks like there are six perspectives which get divided up into what is accesible to the conscious (-1) and the unconscious (-1) and there are two additional perspectives for consciousness (+1). Together they make for an eightfold structure. The key structure is the threesome as it is complemented by a threesome for the unconscious.

DIII ≡ O(16r)/U(8r): The m ∈ m0 ’s are real skew symmetric matrices that anticommute with J1 . We can keep C = ϕ as a particle-hole symmetry and take T = ϕ ⊗ J1 as a time reversal that commutes with H = −iσ2 ⊗ M and squares to −I. The product of C and T is J1 , and this a linear (commutes with −iσ2 ⊗ I) “P ” type symmetry that anticommutes with H.

AII≡ U(8r)/Sp(4r): The generators m ∈ m1 are real skew matrices that commute with J1 and anticommute with J2 . They can be regarded as skew-quaternion-hermitian matrices with complex entries. We no longer need set i → −iσ2 ⊗ I as the matrices no longer have elements coupling between the artificial copies. We instead use J1 as the surrogate for “i.” Now H = J1 m is real symmetric, and commutes with T = J2 . This T acts as a time reversal operator squaring to −I.

CII≡ Sp(4r)/Sp(2r) × Sp(2r): The matrices m ∈ m2 commute with J1 and J2 but anti- commute with J3 . Again H = J1 m. We can set T = J3 as this commutes with with H and squares to −I. P = J2 J3 anticommutes with H but commutes with J1 (and so is a linear map) while C = J2 anticommutes with H, is antilinear and squares to −I.

C≡ {Sp(2r) × Sp(2r)}/Sp(2r) ≃ Sp(2r): The matrices m ∈ m3 commute with J1 , J2 , J3 , and anticommute with J4 , and we can restrict ourselves to the subspace in which K = J1 J2 J3 takes a definite value, say +1. The Hamiltonian J1 m commutes with J4 – but 15J4 does not commute with J1 J2 J3 and so is not allowed as an operator on our subspace. Indeed no product involving J4 is allowed. But C = J2 commutes with J1 J2 J3 and still anticommutes with H. Thus we still have a particle-hole symmetry squaring to −1. The old time reversal J3 now anticommutes with H and looks like another particle- hole symmetry, but is not really an independent one as in this subspace J3 = J2 J1 and J1 is simply multiplication by “i.”

CI≡ Sp(2r)/U(2r): The m ∈ m4 anticommute with J5 . Now J4 J5 commutes with J1 J2 J3 , and so is an allowed operator. It anticommutes with H = J1 m and commutes with J1 . It is therefore a “P ”-type linear map. The map T = J2 J4 J5 is antilinear (anticommutes with J1 ), commutes with H and T 2 = +I. We can take C = J2 again.

AI ≡ U(2r)/O(2r): The m ∈ m5 n anticommute with J6 , and we are to restrict ourselves to the subspace on which K = J1 J2 J3 = +1 and M = J1 J4 J5 = +1. The map T = J3 J4 J6 commutes with K and M and commutes H = J1 m. We have T 2 = +I. We could equivalently take T = J2 J4 J6 .

BDI≡ O(2r)/O(r) × O(r): The m ∈ m6 anticommute with J7 , and we are to restrict our- selves to the eigenspaces of K = J1 J2 J3 , M = J1 J4 J5 . The m also commute with the antilinear operator N = J2 J4 J6 which we can regard as our real structure ϕ. We therefore set C = ϕ = N. Now J3 J4 J7 , J3 J5 J6 , J2 J5 J7 and J1 J6 J7 all commute with K, M and N, each squaring to +I. In the restricted subspace J3 J4 J7 ∝ J2 J5 J7 ∝ J1 J6 J7 and J3 J5 J6 ∝ I. The problem here is what to take for “i,” as the current H → J1 m will take us out of the eigenspace of N. But this is the problem we started with. We need to double the space and keep “i”= J1 and the real structure ϕ = N. We are therefore retaining the R2r Hilbert space. With H = J1 m we have that T = J1 J6 J7 commutes withH and squares to +I.

D≡ {O(r) × O(r)}/O(r) ≃ O(r): Now the m ∈ m7 anticommute with J8 and, except for the factor “i”= J1 , we should stay in the space where K = J1 J2 J3 , M = J1 J4 J5 , N = J2 J4 J6 and P = J1 J6 J7 take the value +1 With H = J1 m we have that C = ϕ = N anticommutes with H, and brings us full circle.

Edit - Upload - History - Print - Recent changes
Search:
This page was last changed on November 11, 2024, at 05:42 PM