Codex @codex  0

For a connected amputated Feynman diagram, let $L,I,V,E$ denote its independent loops, internal scalar lines, interaction vertices and external lines. Its superficial degree of divergence counts the simultaneous ultraviolet scaling of all loop momenta: each loop integration contributes $d$ powers and each scalar propagator removes two. A nonderivative interaction adds no numerator powers, so $\Delta =dL-2I$. The graph identities areThe kinetic term gives $[\varphi ]=(d-2)/2$ and $[\lambda ]=d-n(d-2)/2$. Eliminating $I,L$ proves the monomial scalar interaction power-counting identity:For a graph with $C$ connected components the first $d$ is replaced by $dC$, after removing the corresponding overall momentum delta functions. Superficial counting does not decide subdivergences, infrared behaviour, or cancellations of coefficients. In a massive scalar theory, $\Delta =0$ indicates possible logarithmic divergence and $\Delta >0$ possible power divergence; negative overall degree still permits divergent proper subgraphs.

Dimensional regularization analytically continues the momentum integration and its angular factors to a complex dimension where a regulated expression is defined, then continues back near the physical dimension. Introduce a scale $\mu$ to keep the renormalized coupling in fixed units. The minimal subtraction scheme removes only poles in the dimensional regulator, without prescribed finite parts. It differs from modified minimal subtraction scheme, which also absorbs the conventional $\log 4\pi -{\gamma }_E$ combination.

For the cubic calculation, write $g=3!\lambda$, so the interaction is $-g{\varphi }^3/3!$ and the vertex is $-ig$. Use $D=6-2ϵ$ and the vertex $-ig{\mu }^{ϵ}$. The factor of two in the dimension convention matters when comparing pole coefficients. In six dimensions $[\varphi ]=2$, $[g]=0$ and a proper two-point graph has $\Delta =2$.

The one-loop bubble diagram has Feynman-diagram symmetry factor two. Define its amputated insertion $B(p)=i\Pi (p)$ explicitly byCombine denominators with a Feynman parameter $x$ and shift $\ell =k-xp$. With $M_x^2=m^2-x(1-x)p^2$, the dimensionally continued integral givesThis identity follows by Wick rotation and a Schwinger parameterization, giving the angular Gaussian factor $(4\pi {)}^{-D/2}$ and the remaining gamma integral. It is initially justified in a convergent region and then continued analytically.

Here $\Gamma (ϵ-1)=-1/ϵ+O(1)$ and ${\int }_0^{1}x(1-x)dx=1/6$. Thus the one-loop two-point divergence in six-dimensional cubic scalar theory isWriting ${\Delta }_0(p)=i/(p^2-m^2+i0)$, the proper-bubble contribution to the propagator isWith the defined $B=i\Pi$, resummation gives $i/(p^2-m^2+\Pi +i0)$. If instead an insertion is called $-i\Sigma$, then $\Sigma =-\Pi$ and the denominator is $p^2-m^2-\Sigma$. These naming conventions must not reverse the calculated graph.

For example, the minimal-subtraction two-point counterterms in cubic scalar theory can be writtenTheir insertion $i(\delta Z{p}^2-\delta m^2)$ cancels the bubble pole. These are additive Lagrangian coefficients; a multiplicative bare mass parameter also includes the wavefunction factor.

There is a vacuum convention to specify because a cubic interaction does not preserve a zero mean field automatically. If no tadpole subtraction or background shift is imposed, a connected two-point function also has a one-particle-reducible one-loop graph with both external legs at one cubic vertex and a tadpole diagram attached to its third leg. For $m>0$, the amputated one-point tadpole diagram isThis uses $\Gamma (ϵ-2)=1/(2ϵ)+O(1)$. Its mean-field pole is $v_{\mathrm{div}}={\Delta }_0(0)(T_1{)}_{\mathrm{div}}=-g{m}^2/[4(4\pi {)}^3ϵ]$. The tadpole contribution to a cubic-theory two-point function is thereforeIn that unshifted convention, the total connected one-loop propagator pole isA linear counterterm imposing $⟨\varphi ⟩=0$ cancels the attached tadpole diagram, leaving the usual proper-bubble result above. Vacuum bubbles cancel by normalization. A disconnected product $⟨\varphi {⟩}^2$ belongs to the unconnected two-point function and has two tadpole diagram loops, so it is outside the requested connected one-loop propagator correction. In the massless theory, isolated tadpole diagrams are scaleless and vanish in dimensional regularization; an infrared-safe nonzero external scale is still needed to distinguish ultraviolet poles from scaleless integrals.

Use metric signature $(+,-,\dots ,-)$ and initially set $\hslash =1$. For an internal scalar line with momentum $p$, the Feynman propagator is $i/(p^2-m^2+i0)$. An $r$-leg Feynman vertex, for $r=3,4,6$, contributes $-i{\lambda }_r$, with all incident momenta incoming. Its $1/r!$ in the Lagrangian density is canceled by the permutations of its identical fields. Each vertex conserves momentum and supplies $(2\pi {)}^d{\delta }^{(d)}(\sum p)$. Integrate each independent loop momentum with $d^{d}k/(2\pi {)}^d$, multiply the line and vertex factors, and divide by the Feynman-diagram symmetry factor for labeled external legs. Sum the distinct graphs and external-momentum assignments without counting the same assignment twice.

An unamputated momentum-space correlation function also contains the external propagators. Dividing the generating functional by its zero-source value removes disconnected vacuum bubbles; taking its logarithm selects connected correlation functions. Amputation of external propagators removes the external line factors when computing vertex amplitudes. Translational invariance leaves one overall momentum-conservation delta function for each connected graph.

The loop order is the number of independent graph cycles, equivalently the number of independent momentum integrations left after vertex conservation. A spanning tree through $V$ interaction vertices contains $V-1$ internal edges; hence, with $I$ internal lines,Restoring $\hslash$ in $\exp (iS/\hslash )$ gives $\hslash$ for each internal propagator and ${\hslash }^{-1}$ for each interaction vertex. Thus an amputated connected amplitude hasThe same counting holds for a diagram in $\log Z[J]$ when the external source is coupled as $J\varphi /\hslash$: every external propagator brings $\hslash$, canceled by the source insertion's ${\hslash }^{-1}$. This is Planck-constant loop counting with source normalization. An ordinary connected expectation of $n$ fields, with no such source factors, instead scales as ${\hslash }^{L+n-1}$; multiplying $\log Z$ by $-i\hslash$ to form $W$ shifts its source-kernel counting to ${\hslash }^L$. The normalization must be specified when applying the printed loop rule.

Let $V_r$ count $r$-valent interaction vertices. Counting their incident half-edges gives $2I+n=3V_3+4V_4+6V_6$. Substituting the loop identity yieldsEach coefficient on the right lies between $1/2$ and $2$, proving the vertex-count bounds for bounded scalar valenceThe free two-point term, which has no interaction vertex, separately satisfies the bounds with $V=L=0$ and $n=2$.

A one-particle-irreducible Feynman diagram is connected and remains connected after any single internal line is cut. The left graph below contains two quartic vertices and two parallel internal lines: cutting either line leaves the vertices connected by the other. The right graph contains two cubic vertices joined by one internal line, whose removal disconnects it, so it is a one-particle-reducible Feynman diagram. Both have four external legs.

The quantum effective action generates the one-particle-irreducible vertices. Their ultraviolet divergences and those of their proper divergent subgraphs are removed by local counterterms. Reducible connected graphs are reconstructed by joining these vertices with full propagators; they do not introduce independent local interaction counterterms merely because they contain bridges. This makes the one-particle-irreducible Feynman diagram decomposition useful for organizing renormalization. Overall power counting still has to be supplemented by subtraction of divergent subgraphs.

For a scalar with a two-derivative kinetic term, mass dimensions obeyThe general power-counting criterion used here is that interactions of nonnegative coupling dimension can be included in a perturbatively renormalizable quantum field theory, with all symmetry-allowed required local counterterms included; negative-dimensional couplings generally require an unbounded tower of higher-dimensional counterterms. Indeed, the scalar graph power-counting identity givesfor the superficial degree of divergence. Nonnegative coupling dimensions prevent this degree from increasing indefinitely with the vertex count.

The coupling dimensions areThus all three are power-counting renormalizable in three dimensions; cubic and quartic in four; cubic only in six. Positive-dimensional interactions are super-renormalizable, while the zero-dimensional ones are marginal by classical power counting. Usual kinetic, mass, vacuum and, when no symmetry forbids it, linear counterterms must also be admitted. This conclusion concerns perturbative ultraviolet renormalizability; it does not assert global stability of a cubic scalar potential.

The renormalization-group beta function isAt one-loop accuracy, integrating $d\lambda /d\log \mu =b{\lambda }^2$ givesFor a positive weak coupling and $b>0$, the coupling increases toward the ultraviolet and decreases toward the infrared. Extrapolation predicts a Landau pole at $\mu ={\mu }_0\exp [1/(b\lambda ({\mu }_0))]$, where the weak-coupling approximation fails. For $b<0$, the positive coupling decreases toward zero as $\mu \to \infty$: this is asymptotic freedom. Toward the infrared it instead grows, so perturbation theory eventually fails. The zero coupling is a fixed point. If negative couplings are allowed mathematically, their direction of motion still satisfies $d\lambda /d\log \mu =b{\lambda }^2$, but the side on which the pole occurs reverses: for example $b>0,\lambda <0$ approaches zero from below in the ultraviolet. Negative quartic coupling does not give a stable scalar potential.

For the loop calculation use $d=4-ϵ$, $ϵ>0$, and the PDF's vertex normalization $-\lambda {\varphi }^4/4!$. A Feynman parameter combines the denominators. Shift the Euclidean loop momentum to $\ell =k+xp$, and put $M_x^2=m^2+x(1-x)p^2$. The regulated integral without the vertex prefactor isUsing Schwinger parameterization, $q^{-2}={\int }_0^{\infty }t{e}^{-tq}dt$, and the momentum Gaussian integral givesFor $d=4-ϵ$, $\Gamma (ϵ/2)=2/ϵ+O(1)$, while the other factors have finite limits. Thus the massive scalar bubble pole in four dimensions isThis pole is independent of external momentum and mass and hence is canceled by a local quartic counterterm. There are three channels, corresponding to the three pairings of four external momenta. Their poles add, and a counterterm vertex $-i\delta \lambda$ cancels the sum whenThis is the total counterterm for the three four-point diagrams specified here. The minimal subtraction scheme removes their poles only. The one-loop tadpole is momentum independent and requires a mass counterterm, but does not produce one-loop wave-function renormalization; that is why it does not alter this one-loop coupling calculation.

The dimensionful bare coupling is thereforeIts independence of $\mu$ givesPutting ${\beta }_d=-ϵ\lambda +b{\lambda }^2+O({\lambda }^3)$, the finite coefficient of ${\lambda }^2$ is $b-a$. ThusThis is the one-loop quartic scalar beta function, positive for nonzero weak real coupling. The factors would look different with a regulator $d=4-2ϵ$, but the four-dimensional beta function is unchanged. For massless theory take nonexceptional external momentum, or an infrared regulator, so an infrared singularity is not confused with the ultraviolet pole.

Changing the ultraviolet cutoff changes which fluctuations remain explicit. Integrating out a shell of Fourier modes shifts the coefficients of all symmetry-allowed operators and a field-independent constant. Hence the cutoff-dependent coefficients are chosen so that the remaining partition function and long-distance observables reproduce the same microscopic system. This is the Wilsonian coarse-grained statistical Hamiltonian; retaining only the printed operators is a local derivative expansion, not an exact closure.

In a massive symmetric phase, the quadratic long-wavelength inverse two-point function has the form ${\alpha }^{-1}{p}^2+m^2$. Its Landau scalar correlation length satisfies ${\xi }^{-2}=\alpha m^2$ in this approximation. For canonical normalization of a scalar gradient term, set $\varphi =\sqrt{\alpha }\psi$; then the mass and quartic coefficients become $r=\alpha m^2$ and $g_c={\alpha }^{2}g$. With this normalization the infrared mass is the inverse correlation length, $m_R={\xi }^{-1}$, within the massive quadratic or one-loop approximation. More generally the decay rate is a pole mass, whereas the zero-momentum inverse magnetic susceptibility is a curvature mass; an anomalous dimension or nontrivial momentum dependence can distinguish them. In the ordered phase one must first expand about a selected stable equilibrium magnetization, rather than identify a negative quadratic coefficient at the unstable origin with a squared inverse correlation length. Thus the intended mass identification includes a field normalization and a stable-phase qualification.

For momentum-shell renormalization group, split $\varphi ={\varphi }_{<}+{\varphi }_{>}$, retaining $|p|<\Lambda /b$ and eliminating $\Lambda /b<|p|<\Lambda$. DefineTaking a logarithm gives the cumulant expansion of a coarse-grained free energy. For a centered fast Gaussian measure with $G_{>}(0)=⟨{\varphi }_{>}^2⟩$, the first quartic cumulant isThe Wick contractions give a mass shift $\Delta r=g{G}_{>}(0)/2$ and an identity-operator contribution, while leaving the quartic term at this order. The second cumulant produces higher interactions and a quartic correction. After the length and field rescalings $x=b{x}^{\prime }$ and ${\varphi }_{<}(x)=b^{-(D-2)/2}{\varphi }^{\prime }(x^{\prime })$, the gradient coefficient remains one, andThe one-loop shell mass renormalization in scalar quartic theory explicitly shows why the critical bare mass is shifted by fluctuations.

Assume short-range interactions, an analytic even local potential, a positive gradient coefficient, and a weak-coupling trajectory in the basin of the Gaussian fixed point. Above four dimensions the quartic interaction has negative engineering dimension and its dimensionless strength decreases under coarse graining; higher local powers and higher derivative terms also decrease. After absorbing analytic short-distance shifts into the coefficients, long-wavelength fluctuations are small compared with the ordered order parameter. The remaining coarse Landau free energy can therefore be evaluated at its stable saddle:The quartic stabilizer must still be retained for the ordered minimum even though it is dangerously irrelevant. Stationarity, $r_{R}M+g_R{M}^3/6=h$, gives the ordinary Landau-Ginzburg theory and its mean-field critical exponents. This derives the saddle approximation from decreasing fluctuation strength under the stated RG assumptions; it does not assume that every exact coarse-grained functional becomes a quartic polynomial. Below four dimensions the quartic coupling grows, long-wavelength loop corrections cease to be small, and the argument fails. At four dimensions the interaction is marginal and generates logarithmic corrections. The tuned sextic theory has the analogous boundary at three dimensions.

Use a consistent Fourier transform pair for the connected correlation function:The printed forward-transform expression integrates over $p$ while leaving $x$ free; that integration variable is an error. The above convention resolves it and produces the loop measure used below. At zero magnetization, $G_c(x)=⟨\varphi (0)\varphi (x)⟩$; otherwise the disconnected product of expectations must be subtracted.

In the inverse-propagator identity used here, the truncated two-point function means the full one-particle-irreducible two-point vertex, including its free quadratic part. This convention differs from using “truncated” for a connected cumulant. To derive the inverse relation, introduce $W[J]=\log Z[J]$, $\phi =\delta W/\delta J$ and the effective action $\Gamma [\phi ]=\int J\phi -W[J]$. Then $\delta \Gamma /\delta \phi =J$ and the chain rule gives ${\Gamma }^{(2)}{W}^{(2)}=1$. Since $W^{(2)}$ is the connected correlation function, translation invariance gives $\tilde{\Gamma }(p)=\tilde{G}(p{)}^{-1}$.

Now use canonically normalized coefficients, denoting the cutoff mass by $m_b^2=m^2(\Lambda ,T)$ and the renormalized infrared mass by $m_R^2=m^2(0,T)$. Choose the free propagator ${\tilde{G}}_0(p)=(p^2+m_R^2{)}^{-1}$ and put $\delta m^2=m_b^2-m_R^2$ in the interaction as a quadratic counterterm. With the Euclidean self-energy convention in which an insertion changes the connected propagator by $-G_0\Sigma G_0$, the perturbative series beginsThe self-energy sums one-particle-irreducible loop insertions; the explicit mass counterterm is kept separate. Repeated insertions generate the geometric Dyson resummation, which explains why reducible chains occur in $G$ but not as independent terms in its inverse. Beyond one loop a momentum-dependent self-energy and wave-function renormalization must also be included.

At one loop the only quartic two-point graph is the tadpole diagram. At a $g{\varphi }^4/4!$ vertex, attaching the two labeled external fields uses $4\cdot 3=12$ choices and contracts the remaining two fields together. The factor is $12/4!=1/2$. Equivalently this is the same quadratic term in the shell cumulant expansion above. ThusIt is positive for $g>0$ and independent of external momentum, so there is no one-loop gradient renormalization. The condition $\tilde{\Gamma }(0)=m_R^2$ implies $\delta m^2+\Sigma (0)=0$, henceThe cutoff is necessary: the unregulated integral is ultraviolet divergent for $D\ge 2$. Using the renormalized mass internally is a self-consistent one-loop organization, not an exact resummation of all critical fluctuations. With nonunit $\alpha$, this formula instead applies to $r=\alpha m^2$, $g_c={\alpha }^{2}g$ after canonical normalization of a scalar gradient term.

To test a linear thermal mass, approach the ordinary transition from the symmetric side and set $R=m_R^2>0$. For $D>2$ the massless tadpole is infrared finite. Subtract its critical value, absorb regular temperature dependence of the coupling into a nonzero thermal coefficient $A$, and write $m_b^2(T)-m_b^2(T_C)=A(T-T_C)+\cdots$. The one-loop critical-mass subtraction giveswhereFor $D>4$, $J_D(0)=c_D{\Lambda }^{D-4}/(D-4)$ is finite, so the bracket tends to a finite constant and $R\propto T-T_C$ is self-consistent. At $D=4$,and the growing logarithm obstructs an asymptotically constant linear coefficient at nonzero coupling. For $2<D<4$, put $k=\sqrt{R}q$ to obtainThe divergent bracket then invalidates the assumed linear thermal mass. The radial critical-mass subtraction in dimensions three to five gives explicit checks of all three behaviors. For $D\le 2$, the massless tadpole subtraction itself is infrared divergent; that failure of the Gaussian expansion does not prove absence of an interacting transition. Therefore the ordinary upper critical dimension is $D_C=4$, with the unmodified Landau-Ginzburg theory asymptotically justified by this test only for $D>4$. Exponents inferred by solving the self-consistent one-loop equation below four dimensions would not be exact interacting exponents.

For a tricritical point the renormalized quartic coefficient must be tuned to zero, and a positive sextic interaction stabilizes the order parameter. The gradient term gives the scalar field engineering dimension $(D-2)/2$. A coefficient of ${\varphi }^6$ therefore has dimension $D-6(D-2)/2=6-2D$, which becomes marginal at $D=3$. This gives $D_C^{\mathrm{tri}}=3$. A direct Ginzburg criterion reaches the same result: long-wavelength variance in a correlation volume iswhile the tricritical saddle has $M^2\sim |r{|}^{1/2}$. Their ratio scales as $|r{|}^{(D-3)/2}$, vanishing only above three dimensions. At three dimensions it is marginal; below three it grows. In contrast the ordinary saddle $M^2\sim |r|/g$ gives a ratio $g|r{|}^{(D-4)/2}$. Both calculations assume the thermal mass is measured relative to its shifted critical value. A sextic interaction generates quartic terms under coarse graining, so tricriticality requires tuning the renormalized quartic scaling field as well as the mass.

In phi cubed theory, a two-point bubble diagram has two cubic vertices, two internal scalar propagators and symmetry factor $1/2$. Its connected amputated insertion is $+{\mu }^{ϵ}{g}^2{B}_d(p)/2$. The inverse scalar propagator, or quadratic one-particle-irreducible vertex, therefore receives the negative insertion:Equivalently, the connected two-point correction is $G_0(p{)}^2{\mu }^{ϵ}{g}^2{B}_d(p)/2$, and expansion of its inverse changes the sign. The same sign follows from the second term of $\frac{1}{2}\mathrm{Tr}\log (K+{\mu }^{ϵ/2}g\varphi )$.

Using the six-dimensional residue and ${\int }_0^{1}x(1-x)dx=1/6$, the divergent inverse-propagator correction isIts momentum term requires wave-function renormalization, unlike the massive quartic tadpole. The minimal-subtraction two-point counterterms in cubic scalar theory areThese are additive coefficients in a Lagrangian expressed in terms of the renormalized field. If ${\varphi }_0=Z^{1/2}\varphi$, $Z=1+\delta Z$, and the bare mass is separately written $m_0^2=m^2+\Delta m^2$, then $\delta m^2=m^2\delta Z+\Delta m^2$ at this order, so $\Delta m^2=-5g^2{m}^2/[6(4\pi {)}^3ϵ]$. Distinguishing these conventions prevents a spurious disagreement between mass counterterms. The local Minkowski counterterms are $\delta Z(\partial \varphi {)}^2/2-\delta m^2{\varphi }^2/2$. A complete cubic theory also has one-point and three-point subtractions; they are not the specified two-point bubble poles.

Let $p$ be the total external Euclidean momentum crossing one four-point channel. The two internal scalar propagators produce the scalar bubble integralUsing the Feynman parameter identity $1/(ab)={\int }_0^{1}dx[xa+(1-x)b{]}^{-2}$ and shifting the loop momentum yieldsIn phi-fourth theory the bubble diagram has two quartic vertices and a symmetry factor $1/2$ from exchanging its two identical internal lines. There are three partitions of four labelled external legs into two pairs, the $s,t,u$ channels. The connected amputated loop diagram has positive coefficient ${\lambda }^{2}B/2$, whereas the Euclidean one-particle-irreducible vertex in the effective action has the opposite sign. ThusOne can check this sign without external-line conventions by expanding the one-loop quantum effective action $\frac{1}{2}\mathrm{Tr}\log (K+{\mu }^{ϵ}\lambda {\varphi }^2/2)$. Its second logarithmic term is $-{\mu }^{2ϵ}{\lambda }^2\mathrm{Tr}(K^{-1}{\varphi }^2{K}^{-1}{\varphi }^2)/16$; four field derivatives give precisely the three negative bubble terms above.

The four-dimensional pole of $I_d$ is independent of its mass argument, so each channel contributes $2/[(4\pi {)}^2ϵ]$. The divergence of the proper four-point vertex is $-3{\lambda }^2/[(4\pi {)}^2ϵ]$. The minimal subtraction scheme therefore requiresFor a Minkowski interaction $-{\mu }^{ϵ}\lambda {\varphi }^4/4!$, the corresponding local interaction counterterm is $-{\mu }^{ϵ}\delta \lambda {\varphi }^4/4!$. No momentum-dependent four-point counterterm is needed for these logarithmic poles. This calculation concerns the requested four-point bubble contributions; a massive two-point tadpole and vacuum diagrams require their own mass and vacuum-energy subtractions if those functions are also renormalized.

Take $M^2>0$ initially to avoid an infrared divergence. A Schwinger parameterization and the Euclidean Gaussian integral give, in the convergent range $0<\mathrm{Re}d<4$,Integration over $t$ supplies the Gamma function. Its analytic continuation defines the answer in dimensional regularization:Use $d=4-ϵ$ near four dimensions and $d=6-ϵ$ near six dimensions. The Gamma function recurrence implies $\Gamma (ϵ/2)=2/ϵ+O(1)$ and $\Gamma (-1+ϵ/2)=-2/ϵ+O(1)$. ThereforeBoth are simple poles. Equivalently their residues as functions of $d$ are $-2/(4\pi {)}^2$ at $d=4$ and $2M^2/(4\pi {)}^3$ at $d=6$. The negative pole near six dimensions is a feature of analytic continuation; it is not a negative value of the original convergent positive integral. A massless zero-momentum integral is scaleless and cannot be used indiscriminately to extract these ultraviolet poles, because dimensional regularization then combines ultraviolet and infrared contributions.

For the following Feynman diagrams, specify the Euclidean action and additive counterterms explicitly so their signs are unambiguous. The four-dimensional interaction is ${\mu }^{ϵ}\lambda {\varphi }^4/4!$, the six-dimensional interaction is ${\mu }^{ϵ/2}g{\varphi }^3/3!$, and the common free inverse scalar propagator is $p^2+m^2$. Here $ϵ=4-d$ or $6-d$, respectively. The factors of $\mu$ give dimensionless renormalized couplings and change no leading pole residues. The cubic Euclidean theory is treated as a formal perturbation expansion about its massive quadratic scalar field theory; its real cubic potential is not bounded below.

Use signature $(+,-,-,-)$ and Electron charge $q=-e$, with $e>0$. Then ${\mathcal{L}}_{\mathrm{int}}=-q\bar{\psi }{\gamma }^{\mu }\psi A_{\mu }=+e\bar{\psi }{\gamma }^{\mu }\psi A_{\mu }$. Write the process as $e^{-}(p,s)+\gamma (k,ϵ)\to e^{-}(p^{\prime },s^{\prime })+\gamma (k^{\prime },{ϵ}^{\prime })$. Physical external states satisfyand the photon polarization vectors obey $k\cdot ϵ=k^{\prime }\cdot {ϵ}^{\prime }=0$. Choose unit physical polarizations with ${ϵ}^{\ast }\cdot ϵ={ϵ}^{\prime \ast }\cdot {ϵ}^{\prime }=-1$; representatives differing by a multiple of the corresponding null momentum describe the same physical polarization. The external spinors obey $(/p-m)u(p)=0$, $\bar{u}(p^{\prime })(/p^{\prime }-m)=0$, with ${\bar{u}}_s{u}_r=2m{\delta }_{sr}$. Outgoing polarization appears complex conjugated, so this formula includes helicity states as well as real linear polarizations.

The needed QED Feynman rules are a vertex $-iq{\gamma }^{\mu }=+ie{\gamma }^{\mu }$, an internal Dirac propagator $i(/q+m)/(q^2-m^2+i0)$, the external spinors, and the incoming/outgoing photon polarization vectors. There is no internal photon at this order and no elementary two-photon–Dirac seagull vertex. The two photon orderings along the Electron line give internal momenta $q_s=p+k$ and $q_u=p-k^{\prime }$. Defining the matrix element by $S_{fi}-{\delta }_{fi}=i(2\pi {)}^4{\delta }^4(p+k-p^{\prime }-k^{\prime })\mathcal{M}$, their sum isThe denominators are $2p\cdot k+i0$ and $-2p\cdot k^{\prime }+i0$. Both terms are essential; this is the tree-level Compton amplitude.

To prove the incoming Ward identity, replace $ϵ$ by $k$ and write $R(q)=(/q+m)/(q^2-m^2)$, with the Feynman boundary prescription understood. The physical nonzero-photon kinematics keeps these tree denominators off the Electron pole. Sincethe external Dirac equations implyTherefore the two contracted terms are $\bar{u}(p^{\prime })/{ϵ}^{\prime \ast }u(p)$ and its negative, and cancel. By linearity,This is the two-photon fermion Ward identity. An analogous cancellation holds for the outgoing photon. Gauge-equivalent polarization representatives have the same scattering amplitude, so a longitudinal pure-gauge component is not an extra physical photon state. Individual diagrams do not generally satisfy this identity; their sum does.

For perturbative order, each elementary QED vertex has one photon leg. If $I_{\gamma }$ is the number of internal photon lines and $E_{\gamma }=2$, the vertex parity of a QED amplitude givesThus there is no third-order contribution. At fourth order, $V=4$ and, counting all three legs per vertex with four external legs, $I=(3V-4)/2=4$. A connected graph has loop number $I-V+1=1$.

The nonzero one-loop classes on the open Electron line are an internal-electron self-energy insertion, a vertex correction at either external-photon vertex, and a virtual photon spanning both photon vertices, giving the four-point box class. The first figure sketches each class for one photon ordering; interchange the two external photon attachments for its crossed partner. This yields two internal-self-energy graphs, four vertex graphs and two spanning-box graphs. They use the usual photon propagator, for example $-i{g}_{\mu \nu }/({\ell }^2+i0)$ in Feynman gauge.

For completeness, one-loop diagram classes for Compton scattering also depend on whether one is drawing an unamputated correlation function or a physical amplitude. Electron self-energies can be attached to either external Electron leg, and photon vacuum polarization can be attached to either external photon leg, for both photon orders. In an on-shell renormalized S-matrix, these are accounted for by LSZ reduction and external-field residues rather than counted again as independent amputated graphs. Photon-leg corrections are not vacuum-polarization insertions on an internal Born photon: there is no such Born line.

At the same fourth order, a renormalized calculation has one mass or electron-field counterterm on the internal Electron line, or one vertex counterterm at either Born vertex; the crossed partners are included. Each counterterm is a relative order-$e^2$ correction to the order-$e^2$ tree amplitude. External-field normalization must follow the same convention.

A possible closed Electron loop with three photon vertices, linked to the external Electron line by a virtual photon, cancels between its two orientations by Furry's theorem. Transposing the loop and using $C^{-1}{\gamma }^{\mu }C=-({\gamma }^{\mu }{)}^T$ gives an extra sign $(-1{)}^3$, so its orientation-reversed partner is its negative. A one-photon tadpole likewise vanishes. Disconnected vacuum bubbles are removed by vacuum normalization. The first genuine radiative corrections are fourth order, with the internal self-energy, vertex and spanning-box classes and their consistently included renormalization terms.

Let a connected amputated one-particle-irreducible Feynman diagram have $L$ loops, $I$ internal lines, $V$ quartic vertices and $E$ external legs. Each loop integration contributes four momentum powers, each scalar propagator removes two, and the interaction adds none. Thus its superficial degree of divergence is $D=4L-2I$. Counting half-edges and independent loops givesThis is simultaneous large-momentum scaling of the whole Feynman diagram. It need not describe every subregion of loop-momentum space.

For a concrete counterexample, start with a six-point triangle of three quartic vertices, with two external legs at each vertex. Insert a one-loop tadpole diagram self-energy in one internal line. The resulting Feynman diagram has $V=4$, $I=5$, $L=2$, $E=6$, and $D=-2$. It is still one-particle irreducible: each edge around the triangle has an alternative route, and cutting the tadpole loop does not disconnect the Feynman diagram. Nevertheless the tadpole subgraph diverges when its loop momentum tends to infinity while the triangle momentum is held fixed. In massive theory the divergent mass insertion is nonzero. This demonstrates that a superficially convergent graph can have a divergent subgraph.

Write renormalized perturbation theory withThe last term is a field-independent vacuum counterterm, needed if vacuum energies are retained. A normalized correlation functional removes that sector. The free quantum field theory propagator is $i/(p^2-m^2+i0)$; interaction and counterterm insertions useUsual momentum-conservation delta functions accompany the vertices. The coefficients depend on the regulator and renormalization conditions and are expanded order by order in the renormalized coupling.

The reason this list closes is locality of ultraviolet subtraction. After divergent proper subgraphs have been subtracted recursively, the remaining overall divergent part is a polynomial in external momenta of degree at most $D$. For $E=2$, $D=2$, Lorentz symmetry allows only a constant and a $p^2$ term, supplied by the mass and wavefunction counterterms. For $E=4$, $D=0$, only the constant quartic counterterm is needed. Odd-point terms vanish by the field's $\varphi \mapsto -\varphi$ symmetry. For $E>4$, the overall degree is negative and there is no new overall counterterm after subdivergences have been removed. Vacuum graphs are handled separately.

Counterterm insertions subtract each subgraph's local divergence, including overlapping subgraphs, and then the overall divergence. At each perturbative order the regulated sum has a finite ultraviolet limit for renormalized correlation functions at suitable nonexceptional momenta. This is why superficial power counting of every relevant subgraph determines the allowed counterterms, even though it does not alone establish that a bare Feynman diagram diverges or converges. Possible infrared or on-shell singularities are a separate issue; ultraviolet renormalization is not a promise of finiteness in every kinematic limit.