Discover millions of ebooks, audiobooks, and so much more with a free trial

Only $11.99/month after trial. Cancel anytime.

Gauge Theory and Variational Principles
Gauge Theory and Variational Principles
Gauge Theory and Variational Principles
Ebook289 pages2 hours

Gauge Theory and Variational Principles

Rating: 2 out of 5 stars

2/5

()

Read preview

About this ebook

This text provides a framework for describing and organizing the basic forces of nature and the interactions of subatomic particles. A detailed and self-contained mathematical account of gauge theory, it is geared toward beginning graduate students and advanced undergraduates in mathematics and physics. This well-organized treatment supplements its rigor with intuitive ideas.
Starting with an examination of principal fiber bundles and connections, the text explores curvature; particle fields, Lagrangians, and gauge invariance; Lagrange's equation for particle fields; and the inhomogeneous field equation. Additional topics include free Dirac electron fields; interactions; calculus on frame bundle; and unification of gauge fields and gravitation. The text concludes with references, a selected bibliography, an index of notation, and a general index.
LanguageEnglish
Release dateJan 18, 2013
ISBN9780486151878
Gauge Theory and Variational Principles

Related to Gauge Theory and Variational Principles

Titles in the series (100)

View More

Related ebooks

Physics For You

View More

Related articles

Reviews for Gauge Theory and Variational Principles

Rating: 2 out of 5 stars
2/5

1 rating0 reviews

What did you think?

Tap to rate

Review must be at least 10 words

    Book preview

    Gauge Theory and Variational Principles - David Bleecker

    PRINCIPLES

    CHAPTER 0

    Preliminaries

    n (open sets, continuous functions, the fundamental theorem of calculus, partial derivatives, integration of smooth functions of several variables, etc.). You should be familiar with the use of indices, and realize that sums are taken over repeated indices. If you lack motivation, you should read the introduction to Chapter 1, as well as the Preface.

    Nearly everything in this chapter plays some role in mathematical physics and will be used later in the book. In Section 0.1, we introduce concepts of multilinear algebra such as tensors, metrics, orientations, and star operators. In Section 0.2, we define manifolds and extend the notions of 0.1 to objects on manifolds. Exterior differentiation and codifferentiation of forms on manifolds are defined, as well as the integration of n-forms on n-manifolds. Stokes’ theorem is proved for compactly supported forms on manifolds without boundary. Maxwell’s equations are explicitly written in terms of differential forms in 0.2.22. In Section 0.3, we define Lie groups, Lie algebras, and related notions (e.g., exponential maps and structure constants). The case of the special unitary group SU(n), which is important in physical gauge theories, is discussed in 0.3.12.

    0.1 MULTILINEAR ALGEBRA AND FORMS

    0.1.1 Notation. Let E and F be vector spaces over . Let Ê be the dual space of E, and let T⁰.⁰(E, F) = F. For p, q > 0, Tp·q(E, F) is the space of multilinear functions (F-valued tensors) × E F. We denote Tp,q(E) by Tp,q(E). If u1,..., upE and , then is defined by

    . Let e1,...,en be a basis for E, with ê¹,...,ên the dual basis. Any f Tp,q(E) can be uniquely written as

    for components be the subspace of T⁰, k(E,F) consisting of f such that f(u1,..., uk) is antisymmetric in u1,..., uk E; Λ⁰(E, F) = F and Λk(E) = Λk(E). For ω∈Λk(E),

    where is antisymmetric in the indices il,..., ik.

    0.1.2 Definition. For α ∈ Λi(E) and β∈Λj(E), we define αΛβ∈ Λi+j(E) by

    (αΛβ)(u1,..., ui+j)

    where the sum is over the set of all permutations σ of {1,..., i + j}, and (−1)σ = ±1 is the sign of σ. The components of (α Λ β) are given by

    where

    For α ∈ Λ⁰(E), we set α Λ β=αβ.

    0.1.3 Definition. A metric on E is a g T⁰,²(E) such that g is symmetric and nondegenerate (i.e., if g(u, υ) = 0 for all υ E, then u=0). An orthonormal (o.n.) basis of E relative to g is a basis e1,...,en such that g(ei,ej)=±δij where δij is the Kronecker delta (δij = 1 if i = j; δij = 0 if i j). A volume element of E relative to g is a µ∈Λn(E) given by ê¹ Λ · · · Λên for an o.n. basis e1,...,en; note that êi∈Λl(E) and µ(el,...,en) = 1. Actually µ depends on the choice of o.n. basis, but only by a factor of ±1. A choice of the two possible volume elements is called an orientation of E. If µ is an orientation for E, then a basis υ1,..., υn of E is positively oriented if µ(υ1,..., υn)>0.

    0.1.4 Definition. If g is a metric on E, then there is an induced metric defined as follows. Let υl,...,υn be any basis of E, and let gij = g(υi, υj). Take gij to be the (i, j)th entry of the inverse of the matrix (gij). For α, β∈Λk(E), we define in terms of components (relative to υ1,...,υn) by

    We can verify that is independent of the choice of basis. For α, β ∈ Λ⁰(E), we let .

    0.1.5 Definition. Suppose that we are given a metric g on E and h on F. Then there is a metric (gh) on Λk(E, F) defined as follows. Let f1,..., fm be a basis for F and write α∈Λk(E, F) as α αafa, αa ∈ Λk(E). Then, for α, β ∈ Λk(E, Fwhere hab =h(fa, fb). Again, (gh)(α, β) is independent of the basis.

    0.1.6 Theorem. Let g be a metric on E, and suppose that µ is a volume element of E relative to g. For n = dim E, there is a unique linear isomorphism *: Λk(E) → Λnk(E), such that for all α, β∈Λk(E).

    Proof. For γ∈Λnk(E), define ϕγ: Λk(Eby ϕγ(α)µ = α Λ γ. We can prove that if ϕγ(α) = 0 for all α ∈ Λk(Eϕγ defines a one-to-one linear map Λnk(E) → Λk(E)^. Since

    this map is an isomorphism. Thus, for each β ∈ Λk(E), there is a unique γ∈Λnk(Efor all α ∈ Λk(E). We take *β

    0.1.7 Notation. Let g be a metric on E, and let gij be as in 0.1.4 (relative to the arbitrary basis υl,..., υn of E). For ω ∈ Λk(E) with components , we define

    We write |g| = |det(gij)|. If el,..., en is an o.n. basis for E, let ηij = g(ei, ej)= ±δij, and set (−1)g = det(ηij). If υj aijei, then gij = g(υi, υjakiamjηkm or G = ATηA in terms of matrices. Thus, |det A| = |g|½, and it follows that if υl,..., υn is positively oriented relative to µ. Finally, we let , and note that .

    0.1.8 Theorem. Let g be a metric on E with orientation µ∈Λn(E), and let υl,..., υn be a positively oriented basis of E. For , we have (in the notation of 0.1.6 and 0.1.7)

    (1)

    Proof. Let for all α, β ∈ Λk(E). Now

    where we have used the antisymmetry of a and β

    0.1.9 Theorem. For ω ∈ Λk(E), we have **ω = (−1)g(−1)k(nk)ω.

    Proof. In terms of components relative to some positively oriented basis, we have

    0.1.10 Remark. There is a star operator *: Λk(E, F) → Λnk(E, F) defined componentwise relative to a basis of Fusing * on Λk(E).

    0.2 MANIFOLDS AND TENSOR ANALYSIS

    0.2.1 Definition (Cn-Manifold). Let M be a set, and suppose M is the union of a number of subsets Ui where i ranges over some (possibly infinite) index set I (i.e., M = ∪ iIUi). In : ϕj(Ui ∩ Uj) → ϕi(Ui ∩ Uj) is C∞ (i.e., has continuous partial derivatives of all orders). A subset V ⊂ M is open if ϕi(Ui ∩ V) is open for all i ∈ I. The collection of open sets is called the topology of M relative to {ϕi|i∈I}. We assume that the topology of M is Hausdorff where ik ∈ I, k= 1,2,3,... . Under these assumptions, {ϕi|i∈I} is called an atlas of M. Two atlases are equivalent if their union is an atlas. An equivalence class of atlases is called a differentiable structure on M, and M together with a differentiable structure is called a (C∞) n−manifold, n being the dimension n in some representative atlas is called a chart or coordinate system.

    0.2.2 Definition. If M and N are manifolds, then a function f: M N is called a (C∞) map, provided that (for all charts ϕi: Ui n on M and ψj: Vi m on N: ϕi(Ui f−1(Vjn is C∞. If f has an inverse that is a map, then f is called a diffeomorphism. If is given its usual differentiable structure, then the collection of maps f: M is denoted by C∞(M).

    Figure 1

    0.2.3 Definition. A curve through a point x∈M is a map y: (a, b)→ M (a<0<b) such that y(0) = x. Curves γ1 and y2 through x are called equivalent if (ϕ°γι)′(0)=(ϕ°γ2)’(0) for some (and hence any) chart ϕ: Un with x ∈ U. An equivalence class of curves through x is called a tangent vector at x; the set of all tangent vectors at x is denoted by TxM. We write y’(0) or

    for the vector in TxM determined by γ. Note that TxM has a natural vector space structure. If Yx ∈TxM (say Yx=γ’(0)) and f∈C∞(M), then (is called the derivative of f along Yx, and is denoted by Yx[f].

    0.2.4 Definition. Let TM = Ux∈M TxM. A vector field on M is a function Y: M TM such that Yx ∈ Tx M and (for all f∈C∞(M)) the function Yx[f] is in C∞(M); we denote this function by Y[f] The set of all vector fields on M is denoted by Γ(TM). If Y, Z∈Γ(TM), then [Y,Z] is that vector field such that [Y, Z]x[f]= Yx[Z[f]]−Z x[Y[f]]. We omit the proof of the existence and uniqueness of [Y, Z] (see Kobayashi and Nomizu [1963]). Observe that [Y, Z]= −[Z, Y], and [Y,[Z,W]] + [W,[Y, Z]] + [Z,[W,Y]]=0. The latter relation is called the Jacobi identity.

    0.2.5 Definition. If f: M N is a map and x M, then f*x: TxM → Tf(x)N is the linear function (differential of f at x) f*x(γ’(0)) = (f°γ)’(0) where γ is a curve through x. If ∈Γ(TM) and Y Γ(TN) are such that f*x(Yx)= Yf(x), then we write Y . If Yand , then can be proved. In the event that f is onto, we write this conclusion as f*[Y,Z] = [f* Y, f*Z].

    0.2.6 Definition. Let ϕ: U n be a chart. The coordinate vector fields 1,..., ∂n on U M are defined by

    where ei is the ith standard unit vector in n. Any Y ∈ Γ(TM), when restricted to U, can be expressed as Y ai∂i, where ai C∞(U). If is another chart with U φ, then (for x U ) you may verify that

    where and on . By an abuse of notation, we can write

    Vector fields can be regarded systems of functions (a¹,...,an), defined on coordinate domains, that transform according to the rule above under a change of coordinates.

    0.2.7 Definition. Let Y ∈ Γ(TM) be such that (for each x M) there is a curve γxM through x with for all t . Such a Y is called complete. For t , define ϕt: M M by ϕt(x) = γx(t). We can prove that ϕt is a diffeomorphism, and that ϕs ϕt = ϕs + t for all s, t . The set {ϕt|t } is called the one-parameter group generated by Y. If Z ∈ Γ(TM), then the Lie derivative of Z along Y is the vector field LyZ defined

    Enjoying the preview?
    Page 1 of 1