Bosonic string theory

Bosonic string theory quantizes a relativistic string using only bosonic worldsheet fields. Lorentz-invariant quantization requires 26-dimensional target spacetime and contains a tachyonic ground state.

Keeping the ordinary reparameterization bc ghost system, four embedding bosons require 44 additional real chiral fermions in each chirality to cancel the worldsheet Weyl anomaly. These are internal matter fields, not an assumed locally supersymmetric RNS construction. Their bilinear currents generate the chiral $\mathfrak{so}(44)$ current algebra, with $\left(\genfrac{}{}{0ex}{}{44}{2}\right)=946$ currents. A full closed fermionic model also requires consistent spin-structure and modular projections; the central-charge count alone does not determine these choices.

In the unprojected polynomial Neveu–Schwarz vertex sector, zero fermion insertions give the usual closed-string tachyon and the massless graviton, two-form and dilaton. Two insertions split into a left-right pair ${\psi }^i{\bar{\psi }}^j{e}^{ipX}$ of internal weights $(1/2,1/2)$ and a same-chirality current pair of weight one. The first has $M^2=-2/{\alpha }^{\prime }$ and 44-squared scalar components. The second requires a bosonic derivative on the opposite side for level matching and gives the two sets of 946 massless gauge vectors. Four-fermion scalar-current products belong to a different insertion count. A chosen global projection may retain a subset, so these operators do not by themselves specify a projected string spectrum.

An open string has a worldsheet with boundary. Its endpoint conditions may be Neumann boundary conditions, Dirichlet boundary conditions, or mixtures selected independently in target-space directions.

For an ordinary critical open string with all Neumann boundary conditions, there are 24 independent transverse creation oscillators at each positive mode number $n$. Summing their nonnegative occupation numbers gives one factor $(1-w^n{)}^{-24}$. The coefficient $d_p=[w^p]F(w)$ counts oscillator polarizations at level $p$, excluding continuous center-of-mass momentum and additional endpoint multiplicities. Cauchy coefficient formula gives $d_p=(2\pi i{)}^{-1}\oint F(w)w^{-p-1}dw$.

The Dedekind eta function modular transformation gives $F(e^{-t})\sim (t/2\pi {)}^{12}\exp (4{\pi }^2/t-t)$. The saddle-point approximation for coefficient extraction has $t_0=2\pi /\sqrt{p-1}$, phase $4\pi \sqrt{p-1}$ and curvature $(p-1{)}^{3/2}/\pi$. Its prefactor $t_0^{12}$ supplies $(p-1{)}^{-6}$ and the Gaussian width supplies $(p-1{)}^{-3/4}/\sqrt{2}$. This explains both the exponential growth and the power $-27/4$.

For the Nambu-Goto phase-space action, ${\dot{P}}_m={\partial }_{\sigma }{J}_m$ with $J_m=e{T}^2{X}_m^{\prime }+u{P}_m$. Thus ${\dot{\mathcal{P}}}_m=J_m(\pi )-J_m(0)$. Free endpoint variations force $J_m=0$ individually. A Dirichlet boundary condition allows an external support or D-brane to absorb momentum in the fixed direction.

The spatial integration-by-parts term in the Nambu-Goto phase-space action pairs endpoint displacement with momentum flux $J_m=e{T}^2{X}_m^{\prime }+u{P}_m$. Free endpoint displacements require zero flux. Fixed normal coordinates instead set those variations to zero, while tangential variations retain a Neumann boundary condition. Boundary conditions must also be preserved by the equations of motion. In a boundary-preserving gauge with $u=0$, mixed Neumann boundary conditions and Dirichlet boundary conditions implement a planar D-brane.

Free endpoint variations set the open string flux $e{T}^2{X}_m^{\prime }+u{P}_m$ to zero. In a boundary-adapted temporal gauge, freedom to vary time forces $u=0$ at the end. Nonzero lapse then gives $X^{\prime m}=0$, a Neumann boundary condition.

In a boundary-adapted temporal gauge, a free open string endpoint has $X^{\prime m}=0$. The Nambu–Goto phase-space constraints then imply $P^2=0$, while $\dot{X}=eP$ implies ${\dot{X}}^2=0$. For $X^0=t$, its spatial speed is one. This null boundary is compatible with a timelike interior worldsheet.

In conformal gauge, a free open-string coordinate with Neumann boundary conditions has a cosine expansion. Canonical quantization turns its nonzero Fourier coefficients into string oscillators.

On $0\le \sigma \le \pi$, a coordinate with Neumann boundary conditions at both ends has a constant position, linear time zero mode and cosine oscillators. A coordinate with Dirichlet boundary conditions at both ends has a linear spatial interpolation between endpoint positions and sine oscillators, with no momentum zero mode. Both have integer frequencies. This differs from an ND boundary condition, which applies different conditions at the two ends of the same coordinate and gives half-integer frequencies.

The Fourier kinetic term ${\sum }_{k>0}i{\alpha }_{-k}\cdot {\dot{\alpha }}_k/k$ supplies conjugate pairs, with ${\alpha }_{-k}={\alpha }_k^{\ast }$. Inverting its symplectic form yields the displayed Poisson brackets. Nonzero modes are independent of the center-of-mass pair, while ${\alpha }_0=\sqrt{2{\alpha }^{\prime }}p$ is the momentum zero mode. The imaginary factors are compatible with a real action up to a total derivative.

A transverse open string coordinate has an ND boundary condition when one endpoint obeys a Neumann boundary condition and the other a Dirichlet boundary condition. On a strip of width $\pi$, its nonconstant spatial eigenfunctions are $\cos ((n+1/2)\sigma )$ when the free endpoint is at zero. They solve the endpoint equations because their derivatives vanish at zero and their values vanish at $\pi$. Reversing the endpoints gives sine eigenfunctions with the same frequencies. The fixed endpoint removes the dynamical worldsheet zero mode.

For one free boson, the mixed ND open-string sector has strip vacuum energy $E_0=1/48$. The strip-to-plane transformation relates this energy to the plane Virasoro algebra zero mode by $E_0=h-c/24$, with central charge $c=1$. Hence its ground conformal weight is $h=1/48+1/24=1/16$. For $d$ independent ND coordinates the weights add to $d/16$. At $d=24$, the plane matter ground weight is $3/2$, while the strip oscillator vacuum energy is $1/2$. The plane physical condition $L_0^{\mathrm{plane}}-1=0$ is therefore equivalent to ${\alpha }^{\prime }{p}_{\parallel }^2+N+1/2=0$. The apparent different constants are coordinate/constraint conventions, not different mass spectra.

A closed string has a cylindrical worldsheet without boundary. Periodicity permits independent left- and right-moving oscillator sectors and, in a compact target direction, winding modes.

The left and right transverse oscillator sectors each have generating function ${\prod }_{m\ge 1}(1-q^m{)}^{-24}$. Its coefficient satisfies $d_N\sim 2^{-1/2}(N-1{)}^{-27/4}{e}^{4\pi \sqrt{N-1}}$. Closed-string level matching requires equal levels but leaves their polarizations independent, so $g_N=d_N^2$. With ${\alpha }^{\prime }{M}^2=4(N-1)$, the exponential is $e^{{\beta }_{H}M}$ with ${\beta }_H=4\pi \sqrt{{\alpha }^{\prime }}$. The smooth rest-mass density is proportional to $M^{-26}{e}^{{\beta }_{H}M}$, since $dN/dM={\alpha }^{\prime }M/2$. This distinguishes degeneracy per level from density per unit mass. The polynomial prefactor matters at the boundary of canonical convergence.

A free closed-string embedding coordinate is the sum of center-of-mass zero modes and independent periodic left- and right-moving oscillator expansions.

Covariant quantization keeps target-spacetime Lorentz covariance manifest, imposes canonical commutation relations on every embedding coordinate and removes unphysical states with the Virasoro constraints.

For integer target dimension $2\le d\le 26$, intercept $a=1$ and ordinary nonzero momentum, the old covariant string quantization conditions leave a positive-semidefinite physical inner product. Quotienting its physical null string states removes gauge degeneracy. A DDF operator basis displays $d-2$ positive transverse oscillators; below dimension 26 the remaining physical longitudinal sector can be organized as a unitary Virasoro algebra sector with central charge $26-d$ and highest weight zero. At $d=26$ this extra sector becomes null and the quotient reduces to the 24 transverse oscillators. The result concerns negative-norm states, not the removal of the bosonic tachyon, and it does not itself cancel the worldsheet Weyl anomaly of a subcritical matter theory.

DDF operators are transverse physical string oscillators constructed as contour integrals of weight-one string vertex operators. Choose a null reference momentum $q$ and transverse polarizations ${ϵ}_i$ orthogonal to it and to a reference on-shell momentum. The integrand is proportional to ${ϵ}_i\cdot \partial X{e}^{inq\cdot X}$; its normalization and $q\cdot p$ are fixed so the operators obey the oscillator algebra in the title and commute with the positive-mode Virasoro constraints. For an open-string tachyon seed, one can use $2{\alpha }^{\prime }q\cdot p=1$. Acting with negative modes shifts momentum and raises oscillator level while preserving the physical mass-shell condition. Their transverse Fock space has positive norm and makes the physical content of the no-ghost theorem for the critical bosonic string explicit.

In flat 26-dimensional bosonic string theory, at ordinary nonzero momentum, take physical string states satisfying $L_{n>0}|\psi ⟩=0$ and $(L_0-1)|\psi ⟩=0$. The inner product restricted to these states is positive semidefinite. Quotienting by physical null string states removes the degeneracy; the resulting Hilbert space is represented by the 24 transverse string oscillators. This is a statement about negative-norm matter states, not about removing the auxiliary Faddeev-Popov ghosts. It does not remove the bosonic tachyon.

For $d=26$, $q^2=-2$ and the physical string state $[{\zeta }_{\mu \nu }{\alpha }_{-1}^{\mu }{\alpha }_{-1}^{\nu }+b_{\mu }{\alpha }_{-2}^{\mu }]|p⟩$, the Virasoro constraints give $b_{\nu }=-q^{\mu }{\zeta }_{\mu \nu }$ and $t={\zeta }^{\mu }{}_{\mu }=-2q\cdot b$. Every solution has the displayed decomposition, where $q\cdot A=0$ and $c$ is symmetric, traceless and transverse to $q$. The vector piece is $L_{-1}(2A\cdot {\alpha }_{-1}|p⟩)$; the scalar piece is $t(L_{-2}+3L_{-1}^2/2)|p⟩/10$. These are null string states, leaving the transverse traceless massive tensor as the physical polarization.

Time-coordinate oscillators produce negative norms in the covariant indefinite Fock space. A physical negative-norm state would violate probabilistic unitarity. Such ghosts differ from the anticommuting worldsheet ghost fields introduced by gauge fixing.

Old covariant quantization retains all target-space string oscillators and imposes $(L_0-a)|\psi ⟩=0$ and $L_n|\psi ⟩=0$ for $n>0$. Negative-mode constraints are adjoints, not further annihilation conditions. Quotienting physical null string states removes redundant polarizations.

At $a=1$, let $k=\sqrt{2{\alpha }^{\prime }}p$, $k^2=-2$. The physical scalar $[{\alpha }_{-1}\cdot {\alpha }_{-1}+(D+4)(k\cdot {\alpha }_{-1}{)}^2/10+(D-1)k\cdot {\alpha }_{-2}/5]|p⟩$ has norm $2(D-1)(26-D)/25$. It is positive below 26, null at 26, and a string ghost state above 26. Matching the ordinary transverse spectrum removes this extra scalar at 26.

A polarization obeys $p\cdot ϵ=0$ and $p^2=(a-1)/{\alpha }^{\prime }$. If $a>1$, spacelike momentum allows a negative-norm timelike polarization. Thus ghost absence requires $a\le 1$. At $a=1$ the null string state identification $ϵ\sim ϵ+\lambda p$ leaves $D-2$ polarizations.

A physical string state obeys the Virasoro mass-shell and positive-mode constraints. States differing by a null state represent the same physical polarization.

A spurious string state is a positive-level Virasoro descendant orthogonal to all physical string states.

A null string state is both physical and spurious. It has zero norm and is quotiented from the physical state space.

The physical state space of old covariant quantization is quotiented by its physical spurious states, which are orthogonal to all physical states. In a ghost-free sector these form the radical of the semidefinite inner product. The quotient retains positive physical polarizations.

For $L_0|p⟩=-|p⟩$, the descendant $Z=(L_{-2}+\frac{3}{2}{L}_{-1}^2)|p⟩$ obeys $L_{1}Z=0$ and $L_{2}Z=(D-26)|p⟩/2$. It becomes physical and null at 26 and supplies the scalar quotient needed for the massive spin-two light-cone gauge in string theory spectrum.

String tension is energy per unit length. The Regge-slope parameter ${\alpha }^{\prime }$ is conventionally defined by $T=(2\pi {\alpha }^{\prime }{)}^{-1}$.

The characteristic string length is ${\ell }_s=\sqrt{{\alpha }^{\prime }}$, with $T=1/(2\pi {\alpha }^{\prime })$. Excited string states have masses of order $1/{\ell }_s$ and higher-derivative corrections become important at that scale.

A Regge trajectory is an approximately linear relation between angular momentum and squared mass. A classical rotating open string obeys $J={\alpha }^{\prime }{M}^2$, with quantum effects adding an intercept.

The Regge limit of a scattering amplitude takes the center-of-mass energy invariant $s$ large at fixed momentum-transfer invariant $t$. A leading trajectory produces a power $s^{\alpha (t)}$, with signature and residue factors. For the bosonic Virasoro–Shapiro amplitude in dimensionless string variables, $\alpha (t)=2+2t$. This differs from fixed-angle high-energy scattering.

The Polyakov action introduces an independent worldsheet metric $g_{\alpha \beta }$:Its metric equation makes $g_{\alpha \beta }$ conformal to the induced worldsheet metric, reducing it classically to the Nambu–Goto action.

Replacing the flat target metric in the Polyakov action by a spacetime metric $g_{\mu \nu }(X)$ gives the displayed string sigma model. The embedding fields $X^{\mu }$ map the worldsheet into target spacetime, while $h_{ab}$ is an independent worldsheet metric. Classical worldsheet diffeomorphisms and Weyl transformations permit conformal gauge; quantum consistency additionally requires vanishing sigma-model beta functions.

In conformal gauge and light-cone gauge in string theory $X^{+}=\kappa \tau$, the transverse part of the string sigma model in a spacetime metric reduces toThe omitted longitudinal term is a total time derivative. Each transverse coordinate obeys a massive wave equation. If the spatial period is $L$, the longitudinal momentum is $p^{+}=T\kappa L$ in this normalization. Keeping the gauge slope explicit prevents missing a factor of $p^{+}$ when changing conventions.

Expand every transverse closed string coordinate in a real orthonormal Fourier basis of the spatial circle. The canonical momenta are $p_{Ia}=T{\dot{q}}_{Ia}$, and spatial orthogonality reduces the field Hamiltonian to the displayed independent harmonic oscillators. Canonical quantization gives ${\omega }_a(a_{Ia}^{†}{a}_{Ia}+1/2)$ for each nonzero-frequency mode; a zero-frequency mode retains the free particle term $p_{Ia}^2/(2T)$. The formal zero-point energy requires a regulator; it is not automatically the massless flat-background intercept. Opposite travelling-wave oscillator combinations diagonalize the translation generator and satisfy closed-string level matching.

For a closed string in a string sigma model in an isotropic plane wave, insert $e^{-i{\omega }_n\tau +2\pi in\sigma /L}$ into the transverse wave equation. Periodicity requires integer $n$, giving the displayed dispersion relation. The zero spatial mode is an oscillator of frequency $|m\kappa |$; when this mass vanishes it becomes a free particle mode and needs the position-plus-linear-time expansion.

Varying the independent metric in the Polyakov action sets the worldsheet stress tensor to zero. For a nondegenerate timelike embedding in two worldsheet dimensions, the metric is proportional to the induced worldsheet metric. The conformal factor cancels from the action, which becomes the Nambu–Goto action. This elimination is classical; the quantum measure and worldsheet Weyl anomaly require separate treatment.

Varying the independent worldsheet metric gives $T_{ab}=T[{\partial }_{a}X\cdot {\partial }_{b}X-\frac{1}{2}{\gamma }_{ab}{\gamma }^{cd}{\partial }_{c}X\cdot {\partial }_{d}X]$. Its classical vanishing is the metric equation and produces the Virasoro constraints.

For positive lapse, eliminating momentum gives $(\dot{X}-u{X}^{\prime }{)}^2/(2e)-e{T}^2{X}^{\prime 2}/2$. This is the Polyakov action for ${\gamma }_{ab}=\left(\begin{array}{cc}{u}^2-e^2{T}^2& u\\ u& 1\end{array}\right)$, up to a Weyl transformation. Thus $e=1/T,u=0$ is conformal gauge.

The Polyakov path integral integrates over string embeddings and worldsheet metrics, divided by worldsheet diffeomorphisms and Weyl transformations. Gauge fixing introduces worldsheet ghosts and moduli integrals.

The Polyakov path integral integrates over string worldsheet metrics as well as embedding fields, modulo worldsheet diffeomorphisms and Weyl transformations. Summing surface topologies gives perturbative two-dimensional quantum gravity coupled to the coordinate matter fields. Gauge fixing retains worldsheet moduli and introduces ghosts; a conformal gauge is locally flat, not a global flat metric on every genus. Cancellation of the worldsheet Weyl anomaly makes the critical bosonic version consistent.

Worldsheet moduli parameterize inequivalent conformal structures and marked insertion points after worldsheet diffeomorphisms and Weyl invariance are fixed. At each genus, string perturbation theory integrates over the corresponding moduli space.

The worldsheet conformal Killing group is the connected residual group of conformal automorphisms of the chosen compact Riemann surface. Its infinitesimal vectors satisfy the Conformal Killing equation, or $P_{1}v=0$ for the worldsheet diffeomorphism ghost operator. Its complex dimensions are three on a sphere, one on a torus, and zero for genus at least two. The sphere group consists of Möbius transformations, while the torus's connected group is translation. Additional discrete automorphisms are handled by the moduli quotient rather than counted as continuous zero modes.

After quotienting infinitesimal worldsheet diffeomorphisms and Weyl transformations, choose a metric variation orthogonal to the gauge directions. Orthogonality to Weyl scaling makes its trace zero; the formal adjoint of the worldsheet conformal Killing operator makes its divergence zero. These are transverse-traceless variations of the two-dimensional metric. In a local conformal complex coordinate, their nonzero components satisfy ${\partial }_{\bar{z}}{t}_{zz}=0$ and its conjugate, linking them to holomorphic quadratic differentials.

The torus worldsheet path integral is invariant under the modular group acting on its complex parameter $\tau$. Integrating over the standard fundamental domain of the modular group counts each inequivalent torus once and excludes the short-proper-time ultraviolet region.

A closed-string scattering amplitude is a worldsheet correlation function of external string vertex operators, integrated over their unfixed positions and worldsheet moduli, with the necessary worldsheet ghost fields.

For a stable $n$-point closed-string scattering amplitude on a genus-$g$ surface, let $d_g$ count complex unmarked worldsheet moduli and $k_g$ count complex dimensions of the worldsheet conformal Killing group. Integrating the $n$ vertex positions and moduli, then fixing $k_g$ positions, leaves $d_g+n-k_g$ complex integrations. The punctured Riemann surface moduli dimension identifies this as $3g-3+n$. Faddeev-Popov ghost fields and antighost insertions implement the removed zero modes and the surviving modulus measure; they do not add independently integrated worldsheet coordinates.

The tree-level Polyakov path integral for closed-string tachyons reduces in conformal gauge to their vertex-position integral on the Riemann sphere. Three positions are fixed by the residual Möbius group; the other $N-3$ complex positions are integrated. The embedding-field Gaussian integral produces momentum conservation and the Koba-Nielsen factor.

At genus zero the residual Möbius transformation group fixes three marked insertion points, leaving one complex position to integrate. The corresponding worldsheet ghost fields give $|z_{ij}{z}_{jk}{z}_{ki}{|}^2$ for the three fixed points. For on-shell weight-$(1,1)$ string vertex operators, these ghost factors cancel the coordinate dependence of the fixed-point normalization, including the limit at infinity. The remaining Koba-Nielsen factor for four tachyons is the integrand of the complex beta integral, producing the Virasoro–Shapiro amplitude.

A single crossing-symmetric scattering amplitude has equivalent pole expansions in different exchange channels. Its scattering-amplitude factorization reproduces the string spectrum and couplings in each channel without adding the channel expansions again.

A worldsheet Green function inverts the scalar kinetic operator after its zero modes have been removed. In a locally flat complex coordinate it satisfies $-(\pi {\alpha }^{\prime }{)}^{-1}\partial \bar{\partial }G(z,w)={\delta }^{(2)}(z-w)$ and is $G(z,w)=-({\alpha }^{\prime }/2)\log |z-w{|}^2$ up to harmonic and constant terms.

A string vertex operator inserts an asymptotic string state at a point of the worldsheet. Its conformal weights encode the physical-state and mass-shell conditions.

These insertions are described by the state-field correspondence of a vertex operator algebra; an individual insertion is not the whole algebra.

An integrated closed-string vertex is inserted as $\int d^{2}zV(z,\bar{z})$. In conformal gauge its matter operator must have conformal weights $(1,1)$ so that the insertion is invariant under worldsheet coordinate changes.

At the first massless closed-string level, the matter vertex has the form ${ϵ}_{\mu \nu }:\partial X^{\mu }\bar{\partial }{X}^{\nu }{e}^{ik\cdot X}:$, with null momentum and a polarization tensor transverse in both indices. Its symmetric trace-free, antisymmetric and trace sectors describe the graviton, Kalb–Ramond field and dilaton.

A closed-string polarization tensor specifies the target-spacetime tensor state multiplying $\partial X^{\mu }\bar{\partial }{X}^{\nu }{e}^{ik\cdot X}$. Its decomposition into symmetric trace-free, antisymmetric and trace parts separates the graviton, Kalb–Ramond field and dilaton polarizations.

Changing a massless closed-string polarization by a momentum-longitudinal tensor changes the integrated vertex by a worldsheet total derivative, equivalently by a BRST-exact state. The corresponding target-space transformations are linearized diffeomorphisms and two-form gauge transformations.

A string mass-shell condition relates the target momentum of a vertex operator to its oscillator level. Together with polarization constraints, it makes the matter vertex have the conformal weights required for an integrated insertion.

The bosonic closed-string tachyon vertex is the normal-ordered exponential $:e^{ip\cdot X}:$ integrated over its worldsheet position. Pairwise contractions of such exponentials generate the Koba-Nielsen factor.

A worldsheet diffeomorphism is a smooth reparameterization of the worldsheet coordinates under which $X^{\mu }$ is a scalar and $g_{\alpha \beta }$ is a rank-two tensor.

On a future-directed patch, choose target time as worldsheet time, $X^0=t$. The momentum equation then gives $P^0=1/e$ and $X^{\prime 0}=0$. Freedom to vary target time at an open-string endpoint sets its shift to zero, simplifying the endpoint momentum-flux condition.

Static gauge identifies selected target-space coordinates with worldvolume coordinates, such as $X^0=\tau$ for a string.

For a string, this local static gauge identifies two embedding coordinates with the worldsheet coordinates, $X^0=t$ and $X^1=\sigma$. The remaining coordinates describe a graph over that coordinate plane. Folded embeddings may require more than one patch.

For $X=(t,\sigma ,k(\sigma +t))$, the vertical coordinate velocity is $k$, but its normal projection has magnitude $|k|/\sqrt{1+k^2}<1$. Tangential motion merely changes string labels. The induced worldsheet metric has determinant $-1$ even when constant-label curves become spacelike for $|k|>1$.

With unit tension and $X^0=t$, $X^1=\sigma$, the Virasoro constraints give $P_1=-\Pi {\Phi }^{\prime }$ and $P_0=-\sqrt{(1+{\Phi }^{\prime 2})(1+{\Pi }^2)}$. The reduced energy density is $\mathcal{H}=-P_0$. On the chiral branch $\Pi ={\Phi }^{\prime }$, the Hamiltonian field equation becomes $\dot{\Phi }={\Phi }^{\prime }$.

A Weyl transformation locally rescales a metric tensor, $g\mapsto e^{2\omega }g$, while retaining the same points and coordinates. Applied to a worldsheet metric, it leaves the embedding $X^{\mu }$ unchanged, and the classical Polyakov action is invariant under these Weyl transformations. Applied to a spacetime metric, a Weyl rescaling can relate different frame conventions, such as a string metric and an Einstein metric; the accompanying field normalizations must be transformed as well. It is not itself a coordinate transformation.

Conformal gauge uses worldsheet diffeomorphisms and Weyl transformations to set $g_{\alpha \beta }=e^{2\varphi }{\eta }_{\alpha \beta }$ locally.

After conformal gauge fixing, independent reparameterizations ${\sigma }^{+}\mapsto f({\sigma }^{+})$ and ${\sigma }^{-}\mapsto \tilde{f}({\sigma }^{-})$ remain, accompanied by a compensating Weyl transformation.

Light-cone gauge uses residual conformal transformations to make the target-space coordinate $X^{+}$ linear in worldsheet time. The Virasoro constraints then determine $X^{-}$ from the transverse fields.

This is the string-worldsheet realization of light cone gauge.

On a patch with ${\alpha }_0^{+}\ne 0$, use worldsheet diffeomorphisms and worldsheet supersymmetry to fix ${\alpha }_{n\ne 0}^{+}=0$ and ${\psi }_r^{+}=0$. The Virasoro constraints and superconformal current constraints then determine the longitudinal modes by the displayed transverse bilinears. In the Ramond sector, reduction of the fermionic zero modes includes the zero-mode Dirac constraint; fixing only nonzero modes does not complete the reduction. Eight transverse bosons and eight transverse fermions remain in the critical ten-dimensional RNS string.

Fixing nonzero longitudinal oscillators leaves the center-of-mass symplectic term, transverse oscillator terms, and a zero-mode multiplier imposing the mass-shell condition. On a patch with $p^{+}\ne 0$, further choosing $x^{+}$ as time solves that constraint for $p^{-}=[p_{\perp }^2+(N-a)/{\alpha }^{\prime }]/(2p^{+})$. This distinguishes partial oscillator gauge fixing from the subsequent time choice.

The worldsheet metric equation in conformal gauge sets the stress tensor to zero, $T_{++}=T_{--}=0$. In oscillator language these are the Virasoro constraints.

Critical closed bosonic strings have independent left- and right-moving Virasoro constraints. Positive modes in both sectors annihilate physical kets, while both zero-mode conditions impose the mass shell and closed-string level matching. Quotient by physical null string states, equivalently use the appropriate BRST cohomology. With $L_0={\alpha }^{\prime }{k}^2/4+N$, the result is $N=\tilde{N}$ and $M^2=4(N-1)/{\alpha }^{\prime }$. Negative matter modes are adjoints rather than additional ket-annihilation constraints.

The covariant physical-state conditions impose positive-mode Virasoro constraints on kets and the adjoint conditions on bras, with intercept $a$ in the zero mode. They are a string version of Gupta-Bleuler quantization. Imposing every positive and negative matter mode on the same state would conflict with the Virasoro central extension. In critical bosonic string theory, $a=1$ and the matter central charge is $26$, canceled by the reparameterization ghosts.

With the oscillator symplectic form of an open string, the quadratic generators $L_n={\sum }_k{\alpha }_k\cdot {\alpha }_{n-k}/2$ obey $\{{\alpha }_k,L_n\}=-ik{\alpha }_{k+n}$. Their Poisson brackets close without a central term, so they are first-class constraints. Quantization requires normal ordering and can produce the Virasoro central extension.

The string nonlinear sigma model couples the worldsheet embedding fields to target-space backgrounds such as a metric, a Kalb–Ramond field and a dilaton. Quantum Weyl invariance requires its beta functions to vanish.

The leading massless-sector spacetime action is proportional to $\int \sqrt{-G}{e}^{-2\Phi }[R+4(\nabla \Phi {)}^2-H^2/12]$, with $H=dB$. The backgrounds are the string-frame metric, dilaton, and Kalb–Ramond field. Massive string exchange generates higher-derivative corrections ordered by ${\alpha }^{\prime }$. This perturbative bosonic massless-sector truncation does not remove the tachyon instability.

The string-frame metric is the target metric appearing directly in the kinetic term of the string nonlinear sigma model. With conventional Euclidean dilaton $\phi$, its leading spacetime gravitational action contains $\sqrt{-g^{(s)}}{e}^{-2\phi }(R+4|\nabla \phi {|}^2)$. The dilaton-dependent Ricci coefficient distinguishes it from the Einstein-frame metric. Frame changes are local field redefinitions where the conformal factor is finite and nonzero; the Lorentzian dilaton coupling sign convention fixes which named field is $\phi$.

For $D>2$, rescale the string-frame metric by $g^{(E)}=e^{-4\phi /(D-2)}{g}^{(s)}$. The Weyl transformation of the Ricci scalar makes the coefficient of the Einstein-frame Ricci term independent of the dilaton. Indeed $g^{(s)}=e^{4\phi /(D-2)}{g}^{(E)}$ contributes factors $e^{2D\phi /(D-2)}$ from the volume, $e^{-2\phi }$ from the original coupling and $e^{-4\phi /(D-2)}$ from the Ricci term; their product is one. Derivative terms supply the transformed dilaton kinetic term. The two-dimensional case is excluded from this particular rescaling.

Expand the string embedding map along target geodesics as $X={\exp }_{\bar{X}}Y$. The quadratic Euclidean fluctuation action has a covariant derivative term and a curvature term $-R_{acbd}{Y}^a{Y}^b\partial {\bar{X}}^c\partial {\bar{X}}^d$. The logarithmically divergent coincident contraction is proportional to ${\alpha }^{\prime }{g}^{ab}$; its curvature contraction is the Ricci tensor. A compensating metric counterterm gives ${\beta }_{ab}^g={\alpha }^{\prime }{R}_{ab}+O({\alpha }^{\prime 2})$. This is why the target gravitational equation arises from quantum worldsheet Weyl anomaly cancellation, rather than from the classical worldsheet metric variation. Infrared regulation and removal of equation-of-motion terms by field redefinitions specify the usual one-loop scheme.

The Kalb–Ramond field is an antisymmetric two-form background coupled to the oriented worldsheet area element. Its gauge-invariant field strength is the three-form $H=dB$.

When ${\partial }_{-}$ is invertible, choose $B_{-m}=0$. Residual gauge parameters are exact one-forms and act trivially on $B$. The remaining independent modes are antisymmetric transverse components $B_{IJ}$, satisfying $\square B_{IJ}=0$; longitudinal components obey $B_{+I}={\partial }_{-}^{-1}{\partial }^J{B}_{IJ}$. There are $(D-2)(D-3)/2$ physical polarizations.