Fact-checked by Grok 2 weeks ago

Group extension

In group theory, a group extension provides a method for constructing a new group G from a given N and a H, formalized by a $1 \to N \to G \to H \to 1, where the map from N to G is injective and the map from G to H is surjective with N. This structure captures how G "extends" H by incorporating N as a normal subgroup, allowing the study of more complex groups in terms of simpler building blocks. Group extensions are fundamental in classifying groups and understanding their properties, such as solvability and nilpotency, as solvable groups can be built as towers of abelian extensions, while nilpotent groups arise from iterated central extensions where N lies in the center of G. Key types include split extensions, which correspond to semidirect products when there exists a homomorphism from H to G that splits the sequence, and central extensions, which play a crucial role in cohomology theory for determining equivalence classes of extensions. The classification of all inequivalent extensions of H by N—up to isomorphism preserving the sequence—forms the extension problem, often resolved using group cohomology; for instance, when N is an abelian H-module, the inequivalent extensions are parametrized by the second cohomology group H^2(H, N). Notable examples encompass direct products as trivial split extensions and non-split cases like the quaternion group as a central extension of the Klein four-group by \mathbb{Z}/2\mathbb{Z}, illustrating how extensions reveal symmetries in geometry, physics, and algebra. These constructions extend to broader contexts, such as abelian categories in homological algebra, underscoring their role in modern mathematics.

Basic Concepts

Definition

In group theory, a group extension provides a framework for constructing a larger group from two given groups, where one serves as a normal subgroup and the other as the quotient group. Formally, an extension of a group Q by a group N is specified by a short exact sequence of group homomorphisms $1 \to N \xrightarrow{i} E \xrightarrow{p} Q \to 1, where i is injective (embedding N into E as a subgroup), p is surjective (covering all of Q), and the sequence is exact, meaning \ker p = \operatorname{im} i and this image is normal in E. This structure implies that E has a normal subgroup isomorphic to N, and the quotient E / (\operatorname{im} i) \cong Q. The short exact sequence captures the essential group-theoretic relationships: the injectivity of i ensures no nontrivial elements of N are identified in E, the surjectivity of p guarantees that Q is fully realized as a , and exactness at E confirms that the embedded copy of N precisely accounts for the of the projection to Q. Here, group are functions f: G \to H between groups that preserve the group operation (f(gh) = f(g)f(h)), and the of a homomorphism f is the normal \ker f = \{ g \in G \mid f(g) = e_H \}, where e_H is the identity in H. A fundamental example of such an extension is the E = N \times Q, where i(n) = (n, e_Q) for n \in N and p(n, q) = q for (n, q) \in E, yielding a trivial extension in which elements of N and Q commute. Another representative case is the E = N \rtimes_\phi Q, induced by a \phi: Q \to \operatorname{Aut}(N) specifying an action of Q on N; the multiplication in E is defined by (n_1, q_1)(n_2, q_2) = (n_1 \phi_{q_1}(n_2), q_1 q_2), with i(n) = (n, e_Q) and p(n, q) = q. For instance, both \mathbb{Z}/6\mathbb{Z} and the S_3 realize extensions of \mathbb{Z}/2\mathbb{Z} by \mathbb{Z}/3\mathbb{Z}, with the former being the \mathbb{Z}/2\mathbb{Z} \times \mathbb{Z}/3\mathbb{Z} and the latter the non-trivial \mathbb{Z}/3\mathbb{Z} \rtimes_{\phi} \mathbb{Z}/2\mathbb{Z} where \phi is the inversion action.

Extension Problem

The extension problem in group theory asks: given groups N and Q, determine all groups E (up to ) such that there exists a short $1 \to N \to E \to Q \to 1, where the inclusion N \hookrightarrow E has N in E and the E \twoheadrightarrow Q is surjective with kernel N. This problem seeks to construct and classify such intermediate groups E that "extend" Q by incorporating N as a . Two extensions $1 \to N \to E_1 \to Q \to 1 and $1 \to N \to E_2 \to Q \to 1 are equivalent if there exists an \varphi: E_1 \to E_2 such that the diagram \begin{CD} 1 @>>> N @>>> E_1 @>>> Q @>>> 1 \\ @. @| @V{\varphi}VV @| @. \\ 1 @>>> N @>>> E_2 @>>> Q @>>> 1 \end{CD} commutes, meaning the maps compose equally on the rows. Equivalence thus preserves the structural relationship between N and Q, focusing on isomorphic realizations of the same extension data. For such an extension to exist, there must be a \phi: [Q](/page/Q) \to \Aut(N) induced by conjugation in E: for q \in [Q](/page/Q) and n \in N, lifting q to some \tilde{q} \in E gives \phi(q)(n) = \tilde{q} n \tilde{q}^{-1}, defining an action of [Q](/page/Q) on N compatible with the group operations. In general, solutions to the extension problem are not up to , as multiple non-equivalent E may arise for fixed N and Q; however, holds in special cases, such as the trivial extension where E \cong N \times [Q](/page/Q).

Classification

Trivial Extensions

In group theory, a short exact sequence of groups $1 \to N \xrightarrow{i} E \xrightarrow{\pi} Q \to 1, where N is in E, is called a trivial extension if it splits (i.e., there exists a s: Q \to E such that \pi \circ s = \mathrm{id}_Q) and the induced action of Q on N—given by conjugation in E via the identification of Q with its image under s—is trivial, meaning the homomorphism \phi: Q \to \Aut(N) is the zero map. Such extensions are characterized by the fact that E is isomorphic to the N \times Q, where the group operation is componentwise: (n_1, q_1) \cdot (n_2, q_2) = (n_1 n_2, q_1 q_2), the i sends n \mapsto (n, 1_Q), and the \pi sends (n, q) \mapsto q. Trivial extensions form a subclass of extensions, in which the conjugating action need not be trivial. A key condition for an extension to be trivial is that N lies in of E (ensuring is trivial) and the sequence admits a splitting . In this case, elements of the image of s commute with every element of N. For example, any A arises as a trivial extension of itself by the \{1\}, via $1 \to A \xrightarrow{\mathrm{id}} A \to \{1\} \to 1, which splits canonically.

Split Extensions

In group theory, a split extension of a group Q by a N refers to a $1 \to N \xrightarrow{i} E \xrightarrow{\pi} Q \to 1 that admits a splitting homomorphism s: Q \to E satisfying \pi \circ s = \mathrm{id}_Q. This splitting implies that E contains a isomorphic to Q that intersects N trivially and generates E together with N. Such extensions are precisely those where E is isomorphic to the N \rtimes_\phi Q, with \phi: Q \to \Aut(N) the encoding the action of the image of s on N by conjugation. The is constructed explicitly as the set N \times Q with the twisted (n_1, q_1)(n_2, q_2) = \bigl( n_1 \cdot \phi(q_1)(n_2),\ q_1 q_2 \bigr), where \cdot is the group operation in N. Trivial extensions arise as the special case where \phi is the trivial , reducing to the N \times Q. A representative example is the S_3, which realizes the split extension \mathbb{Z}/3\mathbb{Z} \rtimes \mathbb{Z}/2\mathbb{Z} via the nontrivial action of \mathbb{Z}/2\mathbb{Z} on \mathbb{Z}/3\mathbb{Z} by inversion.

Non-Split Extensions

A non-split extension of a group Q by a N is a short $1 \to N \to E \to Q \to 1 for which there does not exist a s: Q \to E such that the composition with the projection E \to Q is the identity on Q. In such extensions, the middle group E cannot be isomorphic to a semidirect product N \rtimes Q, distinguishing non-split cases from those where a complementary subgroup isomorphic to Q exists within E. A concrete example is the quaternion group Q_8 of order 8, which realizes the non-split extension $1 \to \mathbb{Z}/2\mathbb{Z} \to Q_8 \to \mathbb{Z}/2\mathbb{Z} \times \mathbb{Z}/2\mathbb{Z} \to 1, with \mathbb{Z}/2\mathbb{Z} \times \mathbb{Z}/2\mathbb{Z} denoting the ; no such section s exists here. For contrast, the dihedral group D_4 of order 8 is a split extension of the same quotient by \mathbb{Z}/2\mathbb{Z}, arising as the (\mathbb{Z}/2\mathbb{Z} \times \mathbb{Z}/2\mathbb{Z}) \rtimes \mathbb{Z}/2\mathbb{Z}. The failure of an extension to split is obstructed by non-trivial 2-cocycles associated to the action of [Q](/page/Q) on [N](/page/N+), reflecting the non-existence of a compatible lifting .

Central Extensions

Properties

A central extension of a group Q by a group N is a short exact sequence $1 \to N \xrightarrow{i} E \xrightarrow{\pi} Q \to 1 such that N is contained in the center Z(E) of E. This centrality condition implies that conjugation by elements of E fixes every element of N pointwise, i.e., for all e \in E and n \in N, e n e^{-1} = n. As a result, N must be abelian, since the commutator subgroup [N, N] is contained in [E, N] = \{1\}. The defining centrality of the extension ensures that the induced action \phi: Q \to \Aut(N), given by choosing a lift \tilde{q} \in E with \pi(\tilde{q}) = q and \phi(q)(n) = \tilde{q} n \tilde{q}^{-1} for q \in Q and n \in N, is trivial. These properties highlight the "central" nature of the kernel, distinguishing such extensions from more general ones where the action may be nontrivial. An equivalent characterization of central extensions is that they are precisely those short exact sequences where the image of the inclusion i: N \to E lands in the center Z(E). Regarding derived length, central extensions contribute to the structure of nilpotent groups: a group constructed via a finite sequence of central extensions admits a central series of corresponding length, and if this series reaches the whole group, the group is nilpotent of class equal to the series length. A representative example is the Heisenberg group H_3(\mathbb{Z}), consisting of $3 \times 3 upper-triangular matrices with entries and ones on the diagonal, which forms a central extension $1 \to \mathbb{Z} \to H_3(\mathbb{Z}) \to \mathbb{Z}^2 \to 1. Trivial extensions, such as direct products E = N \times Q, are central extensions featuring both trivial action and splitting.

Cohomology Classification

Group cohomology provides a powerful framework for classifying group extensions. Specifically, for central extensions of a group Q by an N (where Q acts trivially on N), the isomorphism classes of such extensions $1 \to N \to E \to Q \to 1 are in bijective correspondence with elements of the second group H^2(Q, N). Each element in H^2(Q, N) represents an of extensions, where two extensions are equivalent if there is an between them compatible with the projections to Q and inclusions from N. The classification arises from associating to each extension a 2-cocycle. Given a set-theoretic section s: Q \to E with s(1) = 1, the 2-cocycle f: Q \times Q \to N is defined by s(q_1) s(q_2) = \tilde{f}(q_1, q_2) s(q_1 q_2), where \tilde{f}(q_1, q_2) \in N. For the extension to be associative, f must satisfy the 2-cocycle condition: f(q_1, q_2) \, f(q_1 q_2, q_3) = f(q_1, q_2 q_3) \, f(q_2, q_3) for all q_1, q_2, q_3 \in Q, reflecting the trivial action on N. Two cocycles define equivalent extensions if they differ by a 2-coboundary, i.e., f(q_1, q_2) = \phi(q_1) \phi(q_2) \phi(q_1 q_2)^{-1} for some \phi: Q \to N, and the quotient Z^2(Q, N)/B^2(Q, N) = H^2(Q, N) parametrizes the classes. An extension splits, meaning E \cong N \rtimes Q (or direct product for central case), if and only if the corresponding cohomology class in H^2(Q, N) is zero, i.e., the cocycle is a coboundary. This criterion distinguishes split extensions from non-split ones, with non-trivial classes yielding inseparable structures. For general (non-central) extensions, where Q acts non-trivially on N via conjugation in E, the classification still uses H^2(Q, N), but now N is viewed as a Q-module under this action. The cocycle condition generalizes to account for the action: f(q_1, q_2) \, f(q_1 q_2, q_3) = q_1 \cdot f(q_2, q_3) \, f(q_1, q_2 q_3), ensuring compatibility with the module structure, and equivalence classes are again given by H^2(Q, N). A concrete example is the classification of central extensions of the cyclic group \mathbb{Z}/p\mathbb{Z} by \mathbb{Z}/p\mathbb{Z} for prime p, where H^2(\mathbb{Z}/p\mathbb{Z}, \mathbb{Z}/p\mathbb{Z}) \cong \mathbb{Z}/p\mathbb{Z}. The trivial class yields the direct product \mathbb{Z}/p\mathbb{Z} \times \mathbb{Z}/p\mathbb{Z}, while non-trivial classes produce the cyclic group \mathbb{Z}/p^2\mathbb{Z}. For higher-dimensional cases, central extensions of the elementary abelian group (\mathbb{Z}/p\mathbb{Z})^2 by \mathbb{Z}/p\mathbb{Z} are classified by H^2((\mathbb{Z}/p\mathbb{Z})^2, \mathbb{Z}/p\mathbb{Z}) \cong (\mathbb{Z}/p\mathbb{Z})^3, with non-trivial classes yielding extraspecial p-groups of order p^3, such as the Heisenberg group over \mathbb{F}_p, characterized by a non-degenerate alternating bilinear form from the cocycle.

Applications

Lie Groups

In the context of Lie groups, an extension is defined as a short exact sequence $1 \to N \xrightarrow{i} G \xrightarrow{p} Q \to 1, where N, G, and Q are groups, and the maps i and p are smooth group s such that i is injective, p is surjective, and the image of i equals the of p. This structure preserves the aspects of the groups involved, ensuring that G inherits a smooth group operation compatible with the topology and the extension maps. Such extensions are fundamental in for constructing larger groups from smaller ones while maintaining the smooth properties essential for analysis via Lie algebras. Central extensions play a prominent role in , particularly through the concept of universal central extensions, which provide ways to enlarge a Lie algebra by a central ideal in a manner that captures all possible central extensions. For instance, affine Kac-Moody algebras arise as the universal central extensions of loop algebras associated to finite-dimensional Lie algebras, where the loop algebra \mathfrak{g}((t)) (functions from to \mathfrak{g}) is extended centrally by a one-dimensional ideal generated by a central element. This construction is crucial in infinite-dimensional and has applications in and . A finite-dimensional example is the cover \mathbb{Z}/2\mathbb{Z} \to \mathrm{SU}(2) \to \mathrm{SO}(3), where \mathrm{SU}(2) serves as the universal cover of the rotation group \mathrm{SO}(3), with the kernel acting centrally; this extension resolves topological obstructions in representations of rotations, such as in for particles. While an extension of may split (admitting a as a section), the corresponding extension does not always split ly due to topological constraints, even if it splits abstractly as groups. For connected , continuous splitting fails when the extension class lies in a non-trivial element of the second continuous group H^2(Q; N), reflecting mismatches between algebraic and topological structures; examples include certain extensions of compact where the introduces non-trivial . In general, any short of induces a short of their associated , \mathfrak{n} \to \mathfrak{g} \to \mathfrak{q}, via of the smooth maps at the , allowing the structure of the extension to be analyzed independently of global topological issues.

Topological Groups

In the context of topological groups, a group extension is termed topological if the groups A, E, and G are equipped with compatible topologies making them topological groups, and the inclusion i: A \to E and projection \pi: E \to G are continuous homomorphisms. This setup ensures that the extension respects both algebraic and topological structures, allowing for the study of properties like local compactness, connectedness, and the behavior of homomorphisms under continuity constraints. Such extensions arise naturally in settings where groups carry additional geometric or analytic structure, such as Lie groups or spaces in . A key distinction in topological extensions is the notion of continuous splitting, which requires the existence of a continuous s: G \to E serving as a , i.e., \pi \circ s = \mathrm{id}_G. This is stricter than an algebraic splitting, where a mere set-theoretic or algebraic exists without . Continuous splittings may fail even when algebraic ones exist, due to topological obstructions like the of subgroups or incompatibility with . For example, consider the extension $1 \to \mathbb{Q}/\mathbb{Z} \to E \to \mathbb{R}/\mathbb{Z} \to 1, where E = \mathbb{R} \times (\mathbb{Q}/\mathbb{Z}) with the , and the projection \pi(x, y) = x + \mathbb{Z} + y (identifying \mathbb{Q}/\mathbb{Z} with a dense subgroup of \mathbb{R}/\mathbb{Z}). Algebraically, this splits via of representatives for cosets in \mathbb{Q}, but no continuous splitting exists because any continuous would map the \mathbb{R}/\mathbb{Z} to a compact subgroup of E, which contradicts the of \mathbb{Q}/\mathbb{Z} in the . A prominent example of a topological central extension with continuous issues is the universal covering sequence $1 \to \mathbb{Z} \to \mathbb{R} \to S^1 \to 1, where \mathbb{Z} carries the discrete , \mathbb{R} and S^1 (the circle group) have their standard topologies, the sends n \mapsto n, and the is the exponential map t \mapsto e^{2\pi i t}. This is central since [\mathbb{R}, \mathbb{R}] = 0 and \mathbb{Z} lies in the center. Although local continuous sections exist (reflecting the local triviality of the covering), there is no global continuous section, as it would imply a continuous of the compact S^1 into the contractible \mathbb{R} as a complementing \mathbb{Z}, which is impossible due to topological invariants like non-trivial . This extension highlights obstructions in covering spaces, where the failure of global relates to the non-trivial of S^1. Note that smooth versions of such extensions appear in Lie groups, but the topological framework applies more broadly. The classification of extensions, particularly central ones, relies on continuous . The second continuous group H^2_{\mathrm{cont}}(G, A) classifies equivalence classes of topological extensions of G by A up to topological equivalence, where cocycles are continuous maps satisfying the cocycle condition \omega(gh, a) = \omega(g, h \cdot a) \cdot g \cdot \omega(h, a) (for left actions), and coboundaries arise from continuous 1-cochains. For locally compact groups, this is computed using continuous cochains on compact supports or inductive limits. In broader settings, such as paracompact spaces, sheaf of the constant sheaf \underline{A} on the BG provides an alternative classification, capturing topological obstructions via higher derived functors. Applications of topological group extensions appear prominently in algebraic topology, particularly in the study of covering spaces and their relation to fundamental groups. For a path-connected, locally path-connected covering space p: \tilde{X} \to X, the sequence $1 \to \pi_1(\tilde{X}, \tilde{x_0}) \to \pi_1(X, x_0) \to \mathrm{Aut}(p) \to 1 forms a (possibly ) extension of the deck transformation group by the of the cover, where the action is continuous when topologies are considered. This framework classifies connected coverings via subgroups of \pi_1(X), with non-split extensions corresponding to non-trivial .

References

  1. [1]
    Group Extension -- from Wolfram MathWorld
    An extension of a group H by a group N is a group G with a normal subgroup M such that M=N and G/M=H.
  2. [2]
    group extension - PlanetMath.org
    Mar 22, 2013 · The extension problem in group theory is the classification of all extension groups of a given group G G by a given group H H . Specifically ...
  3. [3]
    Some notes on group extensions | What's new - Terry Tao
    Jan 23, 2010 · Definition 1 (Group extension) An extension of a group {H} ... Some standard notions in group theory can be defined using group extensions.
  4. [4]
  5. [5]
    [PDF] Geometric Group Theory - Clara Löh - Universität Regensburg
    Jun 9, 2022 · Definition 1.3.2 (Group extension). Let Q and N be groups. An ... Rotman. An Introduction to the Theory of Groups, fourth edition ...
  6. [6]
    group extension in nLab
    Jun 28, 2025 · A group extension of a group G G by a group A A is third group G ^ \hat G that sits in a short exact sequence, that can usefully be thought of ...
  7. [7]
    [PDF] 2 Cohomology of groups
    So we may think of the semi-direct product as being the “trivial” extension of G by A (similar to split extensions of modules) and ask whether there are ...<|control11|><|separator|>
  8. [8]
    [PDF] Homomorphisms, Quotient Groups, Representations, and Extensions
    exists, namely the direct product K ×Q, which is called the trivial extension. The theory of extensions seeks to classify and construct nontrivial examples.
  9. [9]
    Semidirect Product -- from Wolfram MathWorld
    A "split" extension G of groups N and F which contains a subgroup F^_ isomorphic to F with G=F^_N^_ and F^_ intersection N^_={e}.
  10. [10]
    Group extensions - GroupNames
    Group extensions ; C · C · unique split extension with trivial action(direct product) ; C · C2×C · unique split extension (direct product) ; C · D · unique split ...
  11. [11]
    [PDF] classification of group extensions and h2 - UChicago Math
    Aug 26, 2011 · In other words the extension is split, thus E is the familiar semi-direct product. A o G from group theory. Lemma 3.15. The map Φ is ...
  12. [12]
    [PDF] Semidirect Products
    Indeed, it is easy to see that no primary cyclic group is a semidirect product. EXAMPLE 7.13. Both S, and Z, are semidirect products of Z3 by Z₂. Example 7.13 ...
  13. [13]
    [PDF] Exact Sequences
    In general, there are many extensions of a given abelian group C by another abelian ... is not split, where Q is the quaternion group, V is the Klein four-group, ...
  14. [14]
    [PDF] Chapter 6: Extensions of groups
    Extensions of groups: like doing “all of chemistry for groups.” (a) Groups built from simple extensions (all groups). (b) Groups built from abelian extensions ( ...
  15. [15]
    [PDF] Abstract Algebra: Supplementary Lecture Notes
    1 (S3 is a semidirect product) ... Prove that the automorphism group of Z2 × Z2 is isomorphic to S3. 3. Show that the nonabelian group Z3 Χ (Z2 × Z2) is isomorphic ...
  16. [16]
    [PDF] Central Extensions of Groups
    An extension of G by the group A is given by an exact sequence of. group homomorphisms. 1 −→ A.Missing: abstract | Show results with:abstract
  17. [17]
    [PDF] Central extensions and almost representations - Purdue Math
    Section 3 is devoted to the central extension (1) and the cohomology classes of its pullbacks arising from asymptotic homomorphisms. Section 4 discusses the ...
  18. [18]
    [PDF] Automorphisms of the discrete Heisenberg group
    Specifically, we give an explicit description of a splitting of this group as an extension of GL(2,Z) by Z2. ... presents H as a central extension of Z2 by Z.
  19. [19]
    [PDF] Math 210B. Group cohomology and group extensions
    In this handout, we will see that H2(G, M) is naturally identified with the set of isomorphism classes of such extensions of G by M. But we emphasize that just ...
  20. [20]
    [PDF] The Theory of p-Groups
    We define an extraspecial group to be a p-group for which G/Z(G) is elementary abelian, and Z(G) has order p. Extraspecial groups appear frequently in ...
  21. [21]
    [PDF] arXiv:math/0611431v4 [math.DG] 8 Mar 2012
    Mar 8, 2012 · Abelian extensions. Definition 3.1. An extension of Lie groups is a short exact sequence with smooth. homomorphisms. 1 → A.
  22. [22]
    [PDF] Central extensions of generalized Kac-Moody algebras.
    The Lie algebra in Theorem 3 is the universal central extension of a Lie algebra associated to the matrix a in Kac [3], and is equal to this Lie algebra when ...
  23. [23]
    [PDF] arXiv:math/0101258v1 [math.DG] 31 Jan 2001
    For example consider the central extension. Z2 → SU(2) = Spin(3) → SO(3) of the three dimensional orthogonal group SO(3) by its double cover Spin(3). Here. SU(2) ...
  24. [24]
    ABSTRACTLY SPLIT GROUP EXTENSIONS - Project Euclid
    stractly, but not continuously, split extensions of connected Lie groups arise from the fundamental group of the image group. The question of the existence ...
  25. [25]
    GROUP EXTENSIONS AND COHOMOLOGY FOR LOCALLY ...
    Our cohomology groups in dimension one are classes of continuous crossed homomorphisms, and in dimension two classify topological group extensions of G by A. We ...
  26. [26]
    [PDF] cohomology of topological groups (*) graeme segal - Dmitri Pavlov
    I shall begin by introducing a category of topological abelian groups. § 1. Let Topab denote the category of compactly generated and locally contractible (*) ...