Deep Foundations and Load Transfer

Introduction

Deep foundations transfer structural loads through weak or compressible near-surface soils to stronger soil or rock at depth, or develop sufficient resistance along the embedded shaft. They are selected when shallow footings cannot satisfy requirements for bearing capacity, settlement, uplift, lateral stability, scour protection, or construction constraints. The principal deep-foundation systems are driven piles, bored piles, micropiles, drilled shafts, and pile groups.

The design problem is not simply to calculate a single ultimate load. It requires understanding how load is transferred, how installation changes the ground, how individual elements interact in groups, and how calculated capacity is verified by testing and observation. A satisfactory design must address ultimate limit states, serviceability limit states, structural resistance, geotechnical resistance, durability, constructability, and environmental effects.

Learning objectives

By the end of this chapter, the reader should be able to:

1. Explain shaft resistance, end bearing, uplift resistance, lateral resistance, and load-transfer mechanisms.
2. Distinguish the behaviour and appropriate applications of driven piles, bored piles, micropiles, drilled shafts, and pile groups.
3. Estimate axial compression and uplift capacity using effective-stress, total-stress, and end-bearing approaches.
4. Evaluate group efficiency, settlement, negative skin friction, and construction effects.
5. Interpret static and dynamic pile-load tests and recognize their limitations.
6. Select a suitable foundation system using geotechnical, structural, environmental, and construction criteria.

1. Deep-foundation systems and terminology

A pile is a relatively slender structural member installed or constructed in the ground. A drilled shaft is generally larger in diameter and is formed by excavating and concreting a borehole. The distinction is partly conventional: local standards may define minimum diameters or terminology differently.

Important terms include:

• Shaft resistance or skin friction: shear resistance mobilized at the pile–soil interface.
• End bearing: resistance mobilized beneath the pile toe.
• Ultimate capacity: maximum geotechnical resistance under a defined loading condition.
• Allowable or design resistance: resistance after applying the relevant safety factor or partial factors.
• Critical depth: a depth beyond which the increase in toe resistance or shaft stress is treated as limited in some empirical methods.
• Load-transfer curve: a relationship between displacement and mobilized shaft or toe resistance, commonly denoted t–z for shaft resistance and q–z for toe resistance.
• Plugging: soil enclosed within an open-ended pile moving with the pile, causing behaviour between a plugged and unplugged pile.
• Set-up: increase in pile resistance after driving, often associated with dissipation of excess pore pressure and reconsolidation.

2. Mechanics of load transfer

For an axially loaded pile, the applied load is transferred progressively from the pile to the ground along the shaft and at the toe. At a depth z, the axial force decreases with depth as interface shear is mobilized. In an idealized continuum description,

where Q is axial force in the pile, z is depth measured downward, u is pile perimeter, and τ(z) is mobilized interface shear stress. Integrating over the shaft gives

where Q_s is total shaft resistance and L is embedded length. For a uniform circular pile with perimeter u = πD and approximately uniform unit shaft resistance f_s,

where D is diameter and f_s has units of force per unit area. Toe resistance is

where A_b is the base area and q_b is the unit toe resistance. The nominal ultimate compressive resistance is therefore

This expression assumes that shaft and toe resistances are compatible with the same loading direction and that no separate failure mechanism or structural limit governs. In design, resistance factors or partial factors are applied according to the governing code and design approach.

Under tensile loading, the toe generally contributes little or no resistance unless a specially enlarged or anchored base is present. Uplift resistance is commonly estimated as

where Q_ult,t is ultimate uplift resistance, Q_s,t is upward shaft resistance, W_p is pile self-weight, and W_soil is the weight of soil reliably carried by the pile or pile group. The soil-weight term must not be added unless the failure mechanism justifies it.

The load–settlement response is usually nonlinear. Shaft resistance may mobilize at relatively small displacement, whereas toe resistance often requires larger displacement. Consequently, a pile can carry a substantial working load through shaft resistance while its toe resistance remains only partly mobilized.

3. Axial compression capacity

3.1 Effective-stress method for sands

For a pile in sand, a common effective-stress representation of unit shaft resistance is

where β is an empirical coefficient incorporating interface friction, lateral stress, and installation effects, and σ′_v is the vertical effective stress at the relevant depth. For a layer of thickness ΔL, the shaft resistance is approximated by

The value of β must be selected from a method calibrated to the pile type, density, installation process, and stress range. It should not be treated as a universal soil constant. Driven displacement piles commonly develop greater radial stress than bored piles, while excavation and stress relief may reduce shaft resistance in bored piles.

An alternative interface-friction form is

where K is the horizontal-to-vertical effective-stress ratio at the pile interface and δ is the pile–soil interface friction angle. This equation is useful conceptually, but K is strongly affected by installation and cannot generally be assumed equal to the at-rest coefficient.

Toe resistance in sand may be written in bearing-capacity form as

where N_q is a bearing-capacity factor and σ′_{v,b} is vertical effective stress at the pile toe. Practical design often limits q_b or N_q because idealized bearing-capacity expressions can overpredict resistance for very long piles, compressible soils, layered profiles, or uncertain toe conditions.

3.2 Total-stress method for cohesive soils

For undrained clay, a widely used approach is the α method:

where s_u is undrained shear strength and α is an empirical adhesion factor. The shaft resistance over a length L is

for layers i. Values of α depend on pile type, strength level, stress history, installation disturbance, and whether the clay is normally consolidated or overconsolidated. Driven piles may generate excess pore pressure during installation and later gain capacity through set-up. Bored piles may suffer softening or remoulding at the shaft and base if excavation is prolonged or poorly controlled.

For an undrained toe failure, a simplified expression is

where N_c is commonly taken near 9 for an idealized deep circular foundation in uniform undrained clay, subject to the limitations of the adopted standard and geometry. The total overburden term and the chosen strength definition must be consistent with the design method.

3.3 Layered soils and weak seams

A pile passing through several layers should be divided into geotechnically meaningful intervals. Shaft resistance is summed by layer, but toe resistance is controlled mainly by the material near the base and by the mechanism extending below it. A thin strong layer may not provide full bearing capacity if it overlies weak compressible soil; conversely, a weak seam near the toe may control settlement or punching failure. Borehole logs, cone penetration tests, groundwater data, laboratory strength tests, and installation records should be interpreted together.

3.4 Structural resistance and geotechnical resistance

The pile must be checked as a structural member for compression, tension, buckling where unsupported, bending, shear, fatigue, corrosion, fire, and connection resistance. The smaller of structural and geotechnical design resistance governs. Concrete piles require checks for handling and driving stresses as well as in-service loads. Steel piles require corrosion allowances or protection. Micropile bars and grout–ground bonds require separate structural and bond checks.

4. Uplift and negative skin friction

Uplift can result from wind, seismic overturning, hydrostatic forces, expansive soil, buoyancy, or temporary construction conditions. The shaft interface must mobilize upward shear, which may be reduced by softening, cyclic loading, or groundwater changes. A tension load test is often valuable where uplift is critical.

Negative skin friction occurs when surrounding soil moves downward relative to a pile. It is caused by consolidation beneath new fill, groundwater lowering, surcharge, reconsolidation of sensitive clay, or nearby excavation. The downward shear is an additional load rather than usable resistance. The depth at which relative movement changes from downward soil movement to pile movement is called the neutral plane. The pile design must satisfy

where Q_neg is the accumulated negative skin friction and Q_design,compression is the factored or allowable pile resistance under the relevant code. The pile must also have adequate structural capacity for the combined load. Mitigation includes preloading and waiting, reducing fill thickness, sleeving, bitumen or low-friction coatings, isolation casings, ground improvement, and designing for the dragload. Coatings should not be credited without evidence of long-term performance and constructability.

5. Lateral capacity and combined loading

Lateral pile response is governed by pile flexural stiffness, soil reaction, pile-head fixity, loading direction, cyclic effects, and group interaction. A laterally loaded pile is commonly modeled using p–y curves, in which p is soil reaction per unit length and y is lateral pile deflection. The governing differential equation for a beam on a nonlinear soil foundation is conceptually

where E_p is pile elastic modulus, I_p is second moment of area, y is lateral deflection, z is depth, and p(y,z) is the soil reaction. Boundary conditions represent free-head, fixed-head, or partially restrained pile-head behaviour. This equation is not solved by treating p as a constant; p generally depends nonlinearly on displacement and depth.

Lateral design must check both ultimate soil reaction and serviceability displacement. Near-surface weak layers, scour, liquefaction, cyclic degradation, and eccentricity can dominate. Pile groups experience shadowing: piles behind the leading row may mobilize less soil resistance. Group p–y adjustments or three-dimensional numerical analyses may therefore be required for heavily loaded foundations.

6. Pile groups, efficiency, and settlement

Pile groups are used to support large loads and moments. A pile cap transfers load among elements, but the distribution is not necessarily equal. Edge piles may receive larger axial loads under moment, and piles may be subjected to tension while others are compressed.

A simple preliminary estimate uses group efficiency:

where Q_g,u is estimated group ultimate resistance, η_g is group efficiency, n is number of piles, and Q_1,u is ultimate resistance of one pile. Efficiency is not a universal constant. It depends on spacing, arrangement, pile type, soil, load direction, installation sequence, cap contact, and failure mechanism. Values above or below unity are possible in practice; a value greater than unity should only be used when supported by a suitable method or test.

For closely spaced piles in clay, the group may behave as a block. The block perimeter and base are then checked as a larger foundation, and the smaller resistance from individual-pile action and block action may govern. Group settlement includes elastic compression of piles, shaft-transfer deformation, toe deformation, and compression of the soil mass influenced by the group. It may be much larger than settlement of an isolated pile even when individual pile capacity is adequate.

A useful elastic estimate for pile shortening is

where s_p is pile elastic shortening, Q(z) is axial force at depth z, A_p is pile cross-sectional area, and E_p is pile modulus. The equation assumes axial elasticity and requires a compatible load-transfer distribution. It does not represent total foundation settlement.

For serviceability, engineers commonly combine load-transfer analysis, elastic continuum methods, empirical correlations, and load-test evidence. Settlement predictions should include uncertainty ranges rather than a single apparently precise value.

7. Construction effects

Installation is part of the geotechnical design. Driven piles displace soil laterally and downward, may increase horizontal stress and shaft resistance, and can cause heave or lateral ground movement. Driving can also generate excess pore pressures, vibration, noise, and structural damage. Resistance may increase with time in clay through set-up or decrease through relaxation in some sands and weak rocks.

Bored piles reduce noise and vibration but can lose resistance through stress relief, borehole collapse, groundwater inflow, sediment at the base, necking, inclusions, and segregation during concreting. Casing, drilling fluid, tremie concrete, base cleaning, reinforcement placement, and concrete volume records are essential controls. A concrete pour interrupted below groundwater is particularly vulnerable to defects.

Micropile performance depends heavily on drilling method, casing, grout pressure, grout take, bond length, corrosion protection, and load transfer into competent ground. Pressure grouting can improve bond, but excessive pressure may fracture soil or lift nearby structures. Drilled shafts require inspection of the base and sides where practicable, along with continuous concrete placement and integrity verification.

Construction monitoring should record depth, penetration rate, torque, hammer energy, blows per unit penetration, grout pressure and volume, concrete volume, drilling fluid properties, groundwater, and refusal conditions. Deviations from the design ground model should trigger engineering review rather than automatic acceptance.

8. Load testing and interpretation

8.1 Static axial load tests

A static compression test applies load incrementally or continuously while measuring pile-head load and displacement. A tension test measures uplift displacement, and a lateral test records load, deflection, rotation, and sometimes bending strains. The test arrangement must prevent reaction-system interaction with the test pile and must be designed for the maximum applied load.

A load–settlement curve does not always show a sharp failure point. Interpretation may use a code-defined displacement criterion, a plunging criterion, a Davisson-type offset, a hyperbolic extrapolation, or a specified proof-load requirement. The chosen criterion should be stated before interpretation. Test results are influenced by pile age, loading rate, maintained-load duration, installation sequence, soil variability, and whether the test pile is representative of production piles.

8.2 Dynamic testing

High-strain dynamic testing measures force and velocity near the pile head during impact. Wave-equation analysis uses the recorded signals to estimate static resistance and assess driving stresses. Dynamic testing is valuable for production control and rapid assessment, but it is not identical to a static test. Soil damping assumptions, signal quality, hammer modelling, and set-up effects influence interpretation. Where stakes are high, dynamic results should be correlated with static tests or a well-supported signal-matching procedure.

Low-strain integrity testing can identify changes in impedance associated with major defects, but it does not directly establish axial capacity and may have difficulty detecting defects at great depth or in complex geometries. Cross-hole sonic logging is useful for accessible tubes in large concrete shafts, while thermal integrity profiling can identify concrete-volume anomalies. No integrity method substitutes for sound construction records and engineering judgment.

9. Worked example: compression resistance of a pile in clay

Illustrative example. A circular driven concrete pile has diameter D = 0.40 m and an embedded length of 18 m in uniform undrained clay. Take undrained strength s_u = 60 kPa, adhesion factor α = 0.65, and an undrained toe factor N_c = 9. Ignore the overburden term in the simplified toe expression. Estimate nominal ultimate compression resistance.

The pile perimeter is

The unit shaft resistance is

Thus,

The base area is

The unit toe resistance is

Therefore,

The nominal ultimate resistance is

This is not automatically an allowable working load. The designer must apply the governing code factors, check structural resistance and settlement, assess installation disturbance and set-up, and verify whether the assumed α and toe factor are appropriate. If the pile later experiences dragload, that load must be included separately in the compression design.

10. Applied case: illustrative urban bridge foundation

Background. An illustrative bridge pier is proposed beside an operating hospital. The upper 8 m consists of loose fill and soft clay, underlain by dense sand to 25 m and weathered rock below. Groundwater is shallow, vibration and noise are restricted, and the pier is subjected to compression, uplift during extreme wind, and substantial lateral load.

Problem. Shallow foundations would settle excessively. Driven piles could reach dense sand efficiently, but vibration and noise might disrupt the hospital. Bored piles would reduce vibration but require reliable bore stability and base cleaning. The design must also consider seismic cyclic loading and group settlement.

Analysis. The design team compared steel driven piles, bored piles, and large drilled shafts. Driven piles offered installation records and potentially high shaft resistance, but a trial installation showed unacceptable vibration. Small micropiles were feasible within restricted access but would require many elements and careful group-lateral analysis. A drilled-shaft option provided high lateral stiffness and fewer foundation elements, but its construction risk was greater.

Approach and solution. The selected scheme used large-diameter drilled shafts socketed into weathered rock, with temporary casing through the fill and soft clay. A pilot shaft established excavation stability, base-cleaning procedures, concrete placement rates, and integrity-testing requirements. Static compression and lateral tests were specified on representative test shafts. The design included reduced resistance assumptions for potentially disturbed socket surfaces, structural checks for combined axial force and bending, and a contingency micropile underpinning scheme if a shaft could not be completed.

Outcome. The scheme met vibration restrictions and reduced the number of foundation elements. Monitoring confirmed acceptable ground movement, and integrity testing found no critical defects. The case demonstrates that the theoretically highest calculated capacity is not necessarily the best foundation choice; construction reliability, verification, environmental impact, and consequences of failure are equally important.

Learning points. First, installation method changes soil behaviour and must be included in capacity estimates. Second, lateral serviceability may govern even when axial capacity is large. Third, test planning should begin during concept design. Finally, a robust design includes inspection, acceptance criteria, contingency actions, and communication with affected stakeholders.

11. Design workflow and best practice

A defensible deep-foundation design generally follows this sequence:

1. Define structural actions, combinations, performance criteria, design life, durability, and consequences of failure.
2. Develop a ground model from boreholes, cone penetration tests, laboratory tests, groundwater observations, geophysics, and geological understanding.
3. Screen foundation types against access, vibration, noise, spoil, emissions, nearby structures, and available equipment.
4. Select preliminary dimensions and calculate shaft, toe, uplift, lateral, group, structural, and settlement responses using methods appropriate to each soil layer.
5. Consider negative skin friction, downdrag, scour, liquefaction, seismic loading, corrosion, erosion, and construction-stage loads.
6. Plan trial elements and static, dynamic, and integrity tests in proportion to uncertainty and risk.
7. Specify installation controls, records, tolerances, inspection, acceptance criteria, and nonconformance procedures.
8. Update the design using observed installation data and test results; do not treat the original ground model as infallible.
9. Monitor settlement, lateral movement, vibration, groundwater, and adjacent structures during construction and early service.

Computer programs and artificial intelligence can assist with load-transfer calculations, three-dimensional finite-element analyses, automated pile-driving records, anomaly detection, and uncertainty evaluation. They do not remove the need to verify input data, select appropriate constitutive models, check units, understand boundary conditions, or exercise professional responsibility. An apparently precise numerical result from an unsuitable model is not reliable evidence.

Challenges, limitations, and professional responsibility

Deep-foundation calculations are empirical as well as theoretical. The largest uncertainties commonly arise from spatial soil variability, construction deviation, groundwater, pile–soil interface conditions, interpretation of strength, and extrapolation from one test pile to an entire foundation. Safety factors cannot compensate indefinitely for an incorrect ground model or poor construction.

Engineers must follow applicable building codes, geotechnical standards, environmental permits, occupational-safety rules, and quality-assurance procedures. Ethical practice requires transparent reporting of uncertainty, prompt disclosure of defects, avoidance of unsupported capacity claims, and protection of workers, the public, neighbouring structures, and the environment. Spoil disposal, contaminated groundwater, drilling-fluid management, noise, vibration, and carbon emissions should be addressed during option selection rather than after construction begins.

Chapter Summary

Deep foundations transfer load through shaft resistance, toe resistance, or both. Axial compression capacity is commonly represented as the sum of shaft and toe resistance, but the resistance depends strongly on soil type, installation method, stress history, geometry, and displacement. Effective-stress methods are commonly used in sands, while α methods and undrained bearing approaches are used for cohesive soils, subject to calibration and code requirements. Uplift relies mainly on shaft resistance and must be checked under realistic failure mechanisms.

Pile groups can behave as individual piles or as a block, and group settlement may govern even where individual pile capacity is adequate. Lateral behaviour depends on flexural stiffness, soil reaction, fixity, cyclic loading, and group shadowing. Negative skin friction is a dragload caused by downward soil movement and must be distinguished from usable resistance. Driven, bored, micropile, and drilled-shaft systems each have characteristic construction effects and risks.

Static tests provide direct load–displacement evidence; dynamic and integrity tests are valuable complementary tools but require careful interpretation. The best design integrates ground investigation, analytical calculations, structural checks, construction planning, testing, monitoring, and professional judgment.

Key Terms

Deep foundation: Foundation transferring load to deeper soil or rock through shaft and/or toe resistance.

Driven pile: Prefabricated pile installed by impact, vibration, or pressing.

Bored pile: Cast-in-place pile formed in an excavated borehole.

Micropile: Small-diameter drilled and grouted pile used where access or ground conditions are difficult.

Drilled shaft: Large-diameter cast-in-place deep foundation.

Shaft resistance: Shear resistance developed along the pile–soil interface.

End bearing: Resistance developed beneath the pile toe.

Set-up: Time-dependent increase in resistance after installation.

Negative skin friction: Downward interface shear caused by soil moving downward relative to a pile.

Neutral plane: Depth at which relative soil–pile movement changes direction.

Dragload: Axial load induced by negative skin friction.

Group efficiency: Ratio comparing group resistance with the summed resistance of individual piles.

Block failure: Group failure involving the pile group and intervening soil as one mass.

p–y curve: Relationship between lateral soil reaction and pile deflection.

t–z curve: Relationship between shaft shear transfer and relative displacement.

q–z curve: Relationship between toe resistance and toe displacement.

Static load test: Direct test relating applied load to measured pile displacement.

Dynamic load test: Impact-based test using measured force and velocity signals.

Integrity test: Nondestructive test for detecting significant pile or shaft defects.

Downdrag: Another term for load caused by negative skin friction.

Review Questions

Short Answer Questions

1. Distinguish shaft resistance from end-bearing resistance.
2. Why can a bored pile have different shaft resistance from a driven pile of the same dimensions?
3. What is negative skin friction, and when does it develop?
4. State two limitations of dynamic pile testing.
5. Why can pile-group settlement exceed settlement of an isolated pile?

Long Answer Questions

1. Explain the load-transfer mechanism of an axially compressed pile and discuss the assumptions behind Q_u = Q_s + Q_b.
2. Compare driven piles, bored piles, micropiles, and drilled shafts for an urban construction project.
3. Describe methods for estimating shaft and toe resistance in sand and clay, including their principal limitations.
4. Explain how construction records, static tests, dynamic tests, and integrity tests should be integrated into pile acceptance.
5. Discuss the design implications of negative skin friction, including assessment and mitigation.

Analytical / Application Questions

1. A circular pile has D = 0.50 m, L = 20 m, uniform clay with s_u = 75 kPa, α = 0.55, and N_c = 9. Estimate nominal shaft resistance, toe resistance, and total resistance using the simplified undrained method. State why the result is not automatically an allowable load.
2. A pile group supports a heavily eccentric column load. Explain how load is likely to redistribute among the piles and identify the checks required for compression, tension, lateral response, cap action, and settlement.
3. A proposed driven-pile installation beside a hospital produces excessive vibration. Evaluate the engineering and construction factors involved in selecting between bored piles, micropiles, and drilled shafts, and propose a verification programme.

Multiple Choice Questions with Answers

Question 1

Which expression represents the nominal ultimate compression resistance of a pile when shaft and toe mechanisms act together?

A. Q_u = Q_s − Q_b

B. Q_u = Q_s + Q_b

C. Q_u = Q_s ÷ Q_b

D. Q_u = Q_b only

Correct Answer: B. Q_u = Q_s + Q_b

Question 2

For a circular pile of diameter D and embedded length L with uniform unit shaft resistance f_s, which expression estimates shaft resistance?

A. πD²Lf_s

B. πDLf_s

C. DLf_s ÷ π

D. πD²f_s ÷ 4L

Correct Answer: B. πDLf_s

Question 3

In the α method for undrained clay, unit shaft resistance is estimated as:

A. f_s = βσ′_v

B. f_s = Kσ′_v tanδ

C. f_s = αs_u

D. f_s = N_qσ′_v

Correct Answer: C. f_s = αs_u

Question 4

Negative skin friction is best described as:

A. Upward resistance generated by pile uplift

B. Downward shear caused by soil moving downward relative to the pile

C. Toe resistance lost during static testing

D. Lateral resistance caused by wind loading

Correct Answer: B. Downward shear caused by soil moving downward relative to the pile

Question 5

Which pile type generally produces the least construction vibration?

A. Impact-driven steel pile

B. Impact-driven concrete pile

C. Bored cast-in-place pile

D. Vibratory-driven sheet pile

Correct Answer: C. Bored cast-in-place pile

Question 6

A pile has D = 0.40 m, L = 10 m, and uniform f_s = 30 kPa. Using Q_s = πDLf_s, its shaft resistance is approximately:

A. 38 kN

B. 126 kN

C. 377 kN

D. 942 kN

Correct Answer: C. 377 kN

Question 7

What does a p–y curve primarily relate?

A. Pile toe resistance to toe displacement

B. Shaft shear to axial displacement

C. Lateral soil reaction to pile deflection

D. Concrete strength to curing time

Correct Answer: C. Lateral soil reaction to pile deflection

Question 8

Which statement about group efficiency is correct?

A. It is always exactly 1.0

B. It is independent of spacing and soil type

C. It may differ from unity because of pile interaction and installation effects

D. It applies only to single piles

Correct Answer: C. It may differ from unity because of pile interaction and installation effects

Question 9

A static compression load test directly measures the relationship between:

A. Applied load and pile displacement

B. Concrete temperature and curing time

C. Groundwater level and grain size

D. Hammer energy and steel corrosion

Correct Answer: A. Applied load and pile displacement

Question 10

Which condition can produce pile set-up after driving in clay?

A. Dissipation of excess pore pressure and reconsolidation

B. Immediate removal of all shaft resistance

C. Reduction of effective stress to zero permanently

D. Loss of pile self-weight

Correct Answer: A. Dissipation of excess pore pressure and reconsolidation