Plastic Moment Axial Interaction: P-M Yield Surfaces Guide (2026)

1. Introduction to Axial-Bending Interaction in Plastic Limit Analysis

In structural engineering frameworks, members such as portal frame columns, arch ribs, and high-rise core columns rarely experience pure flexure in isolation. Instead, they sustain significant simultaneous axial compression and bending moments. The study of plastic moment axial interaction establishes how coexisting axial thrust reduces the cross-sectional plastic bending resistance $M_p$ to a reduced plastic moment capacity $M_{pc}$.

                      PLASTIC YIELD SURFACE (P-M INTERACTION)
        P / Py
          1.0 +----------------------------------+
              |                                  |
              | *                                |   Unsafe (Plastic Flow)
              |   *                              |
          n   |     *  <--- P-M Interaction Curve|
              |       *                          |
              |         *                        |
              |           *   Safe (Elastic/Yield)
          0.0 +-------------*--------------------+
             0.0           m = Mpc / Mp         1.0

When evaluating plastic moment axial interaction, the cross-sectional stress field undergoes plastification where a portion of the cross-section is dedicated exclusively to resisting axial compression, leaving only the remaining area to form a flexural resisting couple. The resulting p-m interaction curve defines the bounding plastic yield surface governing the true beam-column plastic capacity.

Understanding these interaction mechanics is essential to preventing premature plastic hinge yielding in heavily loaded building columns and industrial frameworks.

2. Cross-Sectional Stress Mechanics Under Combined P and M

2.1 Neutral Axis Shift Under Coexisting Axial Force

Under pure plastic bending, the Equal Area Axis (EAA) divides the cross-section into equal compressive and tensile areas ($A_c = A_t = A/2$). However, when an axial compression force $P$ acts on the section, compressive equilibrium requires $C – T = P$.

Assuming rigid-perfectly plastic material behavior with yield stress $\sigma_y$:

$$\sigma_y A_c – \sigma_y A_t = P \implies A_c – A_t = \frac{P}{\sigma_y}$$

Because $A_c + A_t = A$ and the full plastic axial capacity is $P_y = A \sigma_y$, we obtain:

$$A_c = \frac{A}{2} \left( 1 + \frac{P}{P_y} ight) = \frac{A}{2} (1 + n)$$

$$A_t = \frac{A}{2} \left( 1 – \frac{P}{P_y} ight) = \frac{A}{2} (1 – n)$$

where $n = P / P_y$ represents the non-dimensional axial load ratio ($0 \le n \le 1$). As axial compression increases, the Plastic Neutral Axis (PNA) shifts toward the tension edge, reducing the area available for tensile resistance.

2.2 Fully Plastic Stress Block Decomposition

The fully plastic stress block under combined $P$ and $M$ can be decomposed into two distinct zones:
1. The Axial Resisting Core: A symmetric central core of depth $2 y_0$ centered about the cross-section axis, subjected entirely to uniform compression $\sigma_y$, providing axial resistance $P = 2 y_0 b \sigma_y$ with zero net moment contribution.
2. The Flexural Resisting Outer Zone: The remaining outer portions of the cross-section, which form a pure bending couple with equal tension and compression forces, providing the reduced plastic moment capacity $M_{pc}$.

      TOTAL STRESS BLOCK         =     AXIAL CORE (P)     +     FLEXURAL COUPLE (Mpc)

      +------------------+             +------------------+             +------------------+
      | -σy (Compression)|             |                  |             | -σy (Compression)|
      | -σy              |             | -σy (Core 2y0)   |             |                  |
   ---|------------------|             | -σy              |          ---|------------------|
      | -σy (Shifted PNA)|             +------------------+             |                  |
      | +σy (Tension)    |             |                  |             | +σy (Tension)    |
      +------------------+             +------------------+             +------------------+

This decomposition provides a direct method to derive exact closed-form interaction equations for any cross-section.

3. Mathematical Derivation of P-M Plastic Interaction Equations

3.1 Solid Rectangular Cross-Sections

Consider a rectangular section of width $b$ and depth $h$.

• Full axial plastic capacity: $P_y = b h \sigma_y$.

• Full plastic moment capacity: $M_p = \frac{b h^2}{4} \sigma_y$.

Let the Plastic Neutral Axis shift by distance $y_0$ from the mid-depth. The central axial core has depth $2 y_0$.
The axial load carried by this core is:

$$P = (b \cdot 2 y_0) \sigma_y \implies y_0 = \frac{P}{2 b \sigma_y} = \frac{h}{2} \left( \frac{P}{P_y} ight) = \frac{h}{2} n$$

The reduced plastic moment capacity $M_{pc}$ is the moment capacity of the total section minus the moment capacity of the central core:

$$M_{pc} = M_p – M_{p,core} = \frac{b h^2}{4} \sigma_y – \frac{b (2 y_0)^2}{4} \sigma_y = \frac{b h^2}{4} \sigma_y \left[ 1 – \left( \frac{2 y_0}{h} ight)^2 ight]$$

Substituting $2 y_0 / h = n$:

$$M_{pc} = M_p \left( 1 – n^2 ight) \implies \frac{M_{pc}}{M_p} + \left( \frac{P}{P_y} ight)^2 = 1.0$$

This fundamental parabolic equation defines the exact p-m interaction curve for solid rectangular sections.

3.2 Structural I-Beam and Wide-Flange Profiles

For a wide-flange I-beam (depth $h$, flange width $b_f$, flange thickness $t_f$, web thickness $t_w$, and web height $h_w = h – 2t_f$), the interaction behavior depends on whether the Plastic Neutral Axis lies within the web or penetrates the flanges.

Let $a_w = A_{web} / A = (h_w t_w) / A$ denote the ratio of web area to total area.

Case 1: Neutral Axis in the Web ($n \le a_w$)

When the axial force is small to moderate ($P \le A_w \sigma_y$), the central axial core of depth $2 y_0$ remains entirely within the web:

$$P = 2 y_0 t_w \sigma_y \implies y_0 = \frac{P}{2 t_w \sigma_y}$$

The reduced plastic moment capacity is:

$$M_{pc} = M_p – \frac{t_w (2 y_0)^2}{4} \sigma_y = M_p – \frac{P^2}{4 t_w \sigma_y} = M_p \left[ 1 – \frac{A^2}{4 t_w Z} n^2 ight]$$

Introducing the non-dimensional parameter $c_w = \frac{A^2}{4 t_w Z}$:

$$\frac{M_{pc}}{M_p} = 1 – c_w n^2 \quad (\text{for } n \le a_w)$$

Because web thickness $t_w$ is small, the curve exhibits a very flat initial slope near $n = 0$, meaning small axial forces cause minimal reduction in major-axis plastic moment capacity.

Case 2: Neutral Axis in the Flanges ($n > a_w$)

When axial thrust is high, the axial core encompasses the entire web and penetrates both flanges. The interaction becomes linear:

$$\frac{M_{pc}}{M_p} = \frac{1 – n}{1 – a_w / 2} \quad (\text{for } n > a_w)$$

3.3 Circular Hollow Sections (CHS) and Solid Tubes

For thin-walled circular hollow sections of mean diameter $d_m$ and wall thickness $t$:

• $P_y = \pi d_m t \sigma_y$

• $M_p = d_m^2 t \sigma_y$

The exact trigonometric interaction relationship is:

$$\frac{M_{pc}}{M_p} = \cos \left( \frac{\pi}{2} \cdot \frac{P}{P_y} ight) = \cos \left( \frac{\pi n}{2} ight)$$

4. The Plastic Yield Surface in Generalized Stress Space

In advanced plastic frame analysis, multi-axis bending ($M_x, M_y$) and axial thrust ($P$) combine to form a three-dimensional plastic yield surface $\Phi(P, M_x, M_y) = 0$.

According to the normality rule of associated plasticity, plastic deformation rate vectors $(\dot{ar\epsilon}_0, \dot{\kappa}_x, \dot{\kappa}_y)$ must be orthogonal to the yield surface:

$$\dot{\mathbf{q}} = \dot{\lambda} abla \Phi$$

This normality condition governs post-yield plastic flow in numerical limit analysis algorithms.

5. Step-by-Step Worked Calculation: Wide-Flange Beam-Column Capacity

To demonstrate the calculation of plastic moment axial interaction, let us analyze a standard European HE 200 B (S355 steel, $f_y = 355\text{ MPa}$) structural steel column subjected to combined axial compression and major-axis bending.

       HE 200 B CROSS-SECTION DIMENSIONS
       * Depth h = 200 mm, Flange width bf = 200 mm
       * Flange thickness tf = 15.0 mm, Web thickness tw = 9.0 mm
       * Cross-sectional Area A = 7810 mm^2 (78.10 cm^2)
       * Plastic Section Modulus Zx = 642.6 x 10^3 mm^3
       * Yield Strength fy = 355 MPa (N/mm^2)

5.1 Section Dimensions and Full Capacities

1. Plastic Axial Capacity ($P_y$):
   $$P_y = A \cdot f_y = 7810\text{ mm}^2 \times 355\text{ N/mm}^2 = 2,772,550\text{ N} = 2772.55\text{ kN}$$

2. Full Plastic Moment Capacity ($M_p$):
   $$M_p = Z_x \cdot f_y = 642.6 \times 10^3\text{ mm}^3 \times 355\text{ N/mm}^2 = 228.12 \times 10^6\text{ N}\cdot\text{ mm} = 228.12\text{ kN}\cdot\text{ m}$$

3. Web Area and Area Ratio ($a_w$):
   * Clear web height: $h_w = h – 2 t_f = 200 – (2 \times 15.0) = 170.0\text{ mm}$.
   * Web Area: $A_{web} = h_w \times t_w = 170.0 \times 9.0 = 1530.0\text{ mm}^2$.
   * Web Area Ratio:
   $$a_w = \frac{A_{web}}{A} = \frac{1530.0}{7810.0} = 0.1959 \approx 0.196$$

5.2 Plastic Neutral Axis Location Under Axial Load

Let the column be subjected to design axial compression $N_{Ed} = 850.0\text{ kN}$.

1. Non-Dimensional Axial Force Ratio ($n$):
   $$n = \frac{N_{Ed}}{P_y} = \frac{850.0}{2772.55} = 0.3066$$
   Since $n = 0.3066 > a_w = 0.1959$, the plastic neutral axis has shifted out of the web and into the column flanges.

2. Flange Penetration Depth:
   * Excess axial load carried by flanges:
   $$\Delta P = N_{Ed} – A_{web} f_y = 850,000 – (1530 \times 355) = 850,000 – 543,150 = 306,850\text{ N}$$
   * Flange area required:
   $$A_{flange,req} = \frac{306,850}{355} = 864.37\text{ mm}^2$$
   * Penetration depth $y_f$ into both top and bottom flanges:
   $$2 b_f y_f = 864.37 \implies 2 (200) y_f = 864.37 \implies y_f = \frac{864.37}{400} = 2.16\text{ mm}$$
   * Total height of axial core $= h_w + 2 y_f = 170.0 + 4.32 = 174.32\text{ mm}$.

5.3 Reduced Plastic Moment Capacity Evaluation

Using the Case 2 interaction formula for $n > a_w$:

$$\frac{M_{pc}}{M_p} = \frac{1 – n}{1 – a_w / 2} = \frac{1 – 0.3066}{1 – 0.1959 / 2} = \frac{0.6934}{1 – 0.09795} = \frac{0.6934}{0.90205} = 0.7687$$

The reduced plastic moment capacity under $N_{Ed} = 850\text{ kN}$ is:

$$M_{pc} = 0.7687 \times M_p = 0.7687 \times 228.12\text{ kN}\cdot\text{ m} = 175.36\text{ kN}\cdot\text{ m}$$

The axial compression of $850\text{ kN}$ ($30.7\%$ of axial squash load) reduces the beam-column plastic capacity by $23.13\%$.

6. Design Code Interaction Standards: AISC 360 and Eurocode 3

Structural design specifications implement conservative piecewise linear approximations of the theoretical plastic moment axial interaction curve to simplify practical design while accounting for member stability.

6.1 AISC 360-22 Chapter H Equations

AISC 360 employs the standard bilinear interaction equations for doubly symmetric members under combined flexure and compression:

• For $\frac{P_r}{P_c} \ge 0.2$:
  $$\frac{P_r}{P_c} + \frac{8}{9} \left( \frac{M_{rx}}{M_{cx}} + \frac{M_{ry}}{M_{cy}} ight) \le 1.0$$

• For $\frac{P_r}{P_c} < 0.2$:
  $$\frac{P_r}{2 P_c} + \left( \frac{M_{rx}}{M_{cx}} + \frac{M_{ry}}{M_{cy}} ight) \le 1.0$$

where $P_r, M_r$ are required design strengths, and $P_c, M_c$ are available design strengths.

6.2 Eurocode 3 (EN 1993-1-1) Section 6.2.9 Provisions

Eurocode 3 permits the full plastic moment capacity $M_{pl,Rd}$ to be used without reduction if the axial load satisfies both:
1. $N_{Ed} \le 0.25 N_{pl,Rd}$
2. $N_{Ed} \le \frac{0.5 h_w t_w f_y}{\gamma_{M0}}$

When these limits are exceeded, the design reduced plastic resistance $M_{N,y,Rd}$ is calculated as:

$$M_{N,y,Rd} = M_{pl,y,Rd} \left[ \frac{1 – n}{1 – 0.5 a_w} ight] \le M_{pl,y,Rd}$$

7. Structural Engineering Wrap-Up and Design Synthesis

Analyzing plastic moment axial interaction is fundamental to ensuring the safety of columns, frames, and beam-columns. By accounting for the Plastic Neutral Axis shift and the reduction of flexural resistance under coexisting axial loads, structural designers prevent catastrophic stability failures and establish accurate limit-state collapse thresholds.

Properly applying interaction equations ensures that frame elements possess both the strength and rotational ductility necessary to maintain overall structural equilibrium.

References & Standards Cited

1. American Institute of Steel Construction (AISC). (2022). Specification for Structural Steel Buildings (AISC 360-22). Chicago, IL: AISC.
2. European Committee for Standardization (CEN). (2005). Eurocode 3: Design of steel structures – Part 1-1: General rules and rules for buildings (EN 1993-1-1). Brussels: CEN.
3. Chen, W. F., & Atsuta, T. (2007). Theory of Beam-Columns, Volume 1: In-Plane Behavior and Design. J. Ross Publishing.
4. Horne, M. R., & Merchant, W. (1965). The Stability of Frames. Oxford: Pergamon Press.
5. Trahair, N. S., Bradford, M. A., Nethercot, D. A., & Gardner, L. (2008). The Behaviour and Design of Steel Structures to EC3. 4th Edition, London: Taylor & Francis.

Frequently Asked Questions (FAQ)

Why does small axial compression cause negligible reduction in major-axis moment capacity for I-beams?

For major-axis bending of I-beams, initial axial compression is absorbed entirely by the slender web ($n le a_w$). Because the web is located close to the neutral axis, its contribution to the overall plastic section modulus $Z_x$ is minimal, resulting in a negligible initial drop in $M_p$.

What is the physical meaning of the Plastic Neutral Axis (PNA)?

The Plastic Neutral Axis is the spatial line across a cross-section that separates fibers yielding in tension ($+sigma_y$) from fibers yielding in compression ($-sigma_y$) under the fully plastic limit state.

How does minor-axis bending differ from major-axis bending under axial thrust?

Under minor-axis bending, the axial thrust is absorbed directly by the wide flanges rather than the web. Because the flanges provide the dominant resistance to minor-axis flexure, minor-axis capacity reduces rapidly following a parabolic interaction curve ($m + n^2 approx 1.0$).

When must second-order P-Delta effects be included in interaction checks?

Second-order $P$-$Delta$ and $P$-$delta$ effects must always be included when axial compression acts on flexible or unbraced frames where lateral drift amplification exceeds $10%$ ($alpha_{cr} 1.1$ in AISC).

Why does AISC use a bilinear interaction equation rather than the exact parabolic curve?

The AISC bilinear equation provides a simple, conservative lower-bound approximation that accounts for member buckling, residual stresses, and geometric imperfections across the entire slenderness spectrum.