Influence of Overprotection  on AC Corrosion.   Analysis of a Real Case

Influence of Overprotection on AC Corrosion. Analysis of a Real Case

Ivano Magnifico, Certified Senior Technician in Cathodic Protection.

Meet the Author
Ivano Magnifico holds a master’s degree in Electronic Engineering and serves as the Gas and Oil Product Manager at Automa, an ICorr Corporate Member.

A certified Cathodic Protection Specialist, he combines technical competence with deep knowledge of market analysis and industry standards. With over 15 years of experience in remote cathodic protection monitoring and a patent for an intelligent reference electrode, Ivano has made significant contributions to the field. He is a member of the Board of Directors of CEOCOR (European Committee for the Study of Corrosion and Protection of Piping Systems) and serves as the Delegate of the AMPP Italy Ivano Magnifico Chapter. In addition, he is an active participant in ISO and AMPP standard working groups on cathodic protection.

Introduction

The risk of AC corrosion has always been linked to the parallelisms of underground pipelines with High Voltage AC lines, especially in those geographical areas where the morphology of the territory creates obligatory so-called “technological corridors” and therefore forces the coexistence of different services over long distances.

Recently, the greater diffusion of AC-powered railway networks has further increased the AC interfering sources, while the use of more performing coatings on underground pipelines has on the one hand increased their insulation from the surrounding soil, and on the other has increased the risk of overprotection compared to old, less performing, or more degraded coatings.

This paper, starting from a real case found in a gas distribution network, will present the normative criteria to be used to keep
the AC corrosion risk under control, and will highlight how the simultaneous presence of cathodic overprotection may result in an autocatalytic cycle leading to accelerated AC corrosion, in which monitoring becomes essential in order to be able to carry out on time the appropriate corrective actions.

There are several mechanisms through which an AC source can interfere with a metal structure (Fig. 1): by inductive coupling,
as an effect of the magnetic field generated with respect to an underground structure; by capacitive coupling, in the case of an aerial structure; and by conductive coupling in the presence of a fault current in the ground, in the case of an underground pipeline.

In the case of underground pipelines, under normal operating conditions, the mechanism that can generate AC interference is inductive coupling: normally the interference effect is greater as larger the length of the sections where the pipeline and the AC source (high voltage AC lines, railways operated in AC) follow a parallel path.

AC Corrosion Protection Criteria

International industry standards specify which electrical parameters shall be monitored and their maximum allowed values. The standard ISO 18086:2019 “Corrosion of metals and alloys – Determination of AC corrosion – Protection criteria” indicates two steps for the verification of permissible AC interference levels (Fig.2):

Figure 1: AC Interference Mechanism.

Figure 2: AC Corrosion Risk Assessment According To ISO 18086.

The first step relates to a safety criterion for maximum permissible touch voltage (15V threshold) and does not have a direct rule in AC corrosion risk assessment. This value considers a hand-to-hand or hand-to-foot resistance for an adult male human body of 1500 Ω, yielding a current flow of 10 mA when 15 V is applied [2].

The criterion is based on current density measurements carried out through a coupon whose surface is defined by the standard to be 1 cm², connected to the structure. Both AC current density and DC current density must be measured, as the level of cathodic protection can affect the AC corrosion phenomenon.

NACE standard SP21424-2018 “Alternating Current Corrosion on Cathodically Protected Pipelines: Risk Assessment, Mitigation, and Monitoring” [3] expresses similar values, where depending on the measured DC current density (J.dc) value, different levels of AC current density (J.ac) are allowed:

• If J.dc > 1 A/m2 then J.ac < 30 A/m2; or
• If J.dc < 1 A/m2 then J.ac < 100 A/m2

This standard imposes a maximum AC current density limit even if the DC current density is less than 1 A/m², while the coupon surface of 1 cm² is indicated as generally used but not mandatory.
The Spread Resistance is the ohmic resistance through a coating defect towards remote earth and controls the DC (Idc) or AC (Iac) current passing through a defect at a given voltage (Udc or Uac):
Uac = R’s Iac or Uac = Rs J.ac (1)

where Rs is the normalized Spread Resistance expressed in Ω·m2.

On coating defects, where cathodic protection current reaches the steel surface, cathodic reactions occur involving oxygen reduction and hydrogen evolution. Both reactions generate hydroxide ions(OH-) leading to increased pH at the interface and alkalinity.

Since Spread Resistance depends [4] on both defect size (decreases as surface decreases) and pH value at the interface (decreases as pH increases), the DC current density reaching the defect affects it:

Lower current density leads to decreased pH value and increased Spread Resistance.

Higher current density leads to increased pH value and decreased Spread Resistance.

This is where overprotection can have an effect on AC corrosion:

Presence of a very electronegative IR-free potential (due to high DC current densities);

• Decrease in the Spread Resistance value;

• Possibility of significant AC current density even with low measured AC voltage.

Regarding the choice about which size of coupon to use, increasing the surface area of the coupon results in a lower average current density since the spread resistance increases linearly with increasing defect diameter and the current density decreases linearly with surface area.

Therefore, the current density is typically underestimated when the surface area of the coupon is chosen to be larger than the maximum defect size on the structure: for this reason, in the case of AC corrosion, the standards indicate the use of a 1 cm2 coupon.

AC Corrosion Mechanism in the Presence of Over-Protection [3]

For pipelines with applied cathodic protection, AC corrosion development requires simultaneous coexistence of induced AC, excessive cathodic protection, and small coating defects. Under these conditions:

1. Induced alternating current leads to alternating current discharge on coating defects.

2. AC current density is regulated by alternating voltage and spread resistance associated with the coating defect, through Ohm’s law.

3. Spread resistance depends on:
a. Coating defect size.
b. Soil resistivity near the defect.
c. Soil chemistry.
d. Cathodic protection current density in the coating defect.

Figure 3 – Autocatalytic Nature of AC Corrosion on Cathodically Protected Pipelines Described by Sp21424.

As shown in Fig.3, the AC current density can lead to the depolarisation of the defect: this requires a higher DC current density to maintain a certain cathodic protection potential. Increasing the level of cathodic protection to mitigate AC corrosion, in this case, has the opposite effect: the increase in DC current density further decreases the Spread Resistance at the coating defect due to the production of OH- ions (alkalinisation). Through high levels of cathodic protection, the Spread Resistance decreases, thus increasing the density of alternating current, restarting the cycle: this scenario results in an autocatalytic cycle leading to AC corrosion.

It therefore becomes clear that, in order to leave this cycle, it is necessary to control both the AC current density and the DC current density.

Analysis of A Real Field Case

The case that will be shown has been detected on a measurement point of the distribution network of a large European city, with the following features:

• An extensive cathodic protection system forming a ring around the city center with radiating offshoots.

• Multiple crossings with DC powered railways and surface metro.

• Multiple parallels with the HVAC network.

• Cathodic Protection guaranteed by two T/Rs.

The analysed measuring point (MP):

• Located in a CP system area with several km of parallelism with HVAC line.
• Local soil resistivity between 25 and 50 Ω·m.
• Equipped with permanent CSE reference electrode with integrated 10 cm² coupon (measured current density is underestimated compared to 1 cm² coupon).
• Equipped with a G4C-PRO remote monitoring device capable of performing instant-off measurements on coupon and current density measures.

The measurements shown in Table 1 correspond to daily reports calculated on measurements performed continuously at a frequency of 1 Hz (1 measure per second) for each measuring channel. The minimum, average and maximum daily values are shown over a period of 4 days:

• Eon.dc: ON potential (DC) expressed in V CSE;
• Eon.ac: ON potential (AC) expressed in V;
• E off: instant-off on coupon, equivalent to IR-Free potential
(measured, every second, after a 1 ms wait from switch opening and over a 20 ms interval) expressed in V CSE;
• mIon: DC polarisation current of the coupon expressed in mA; as the coupon size is 10 cm2 the shown value corresponds to the current density in A/m2;
• mIon.ac: AC polarisation current of the coupon expressed in mA; as the coupon size is 10 cm2 the shown value corresponds to the current density in A/m2 (note: the current density value measured on a 1 cm2 coupon would be significantly greater than that measured on the 10 cm2 coupon).

In the absence of coupons, the only available measures would be Eon.dc and Eon.ac, and, on these values, the only possible evaluation would be that relating to the first step of ISO 18086, which would be absolutely respected considering that the highest AC average value along the four days shown (0,424 V) is well below the indicated threshold of 15V. Generally, such a low AC voltage value would never suspect a real risk of AC corrosion, but as can be detected from the DC and AC current densities, we are faced with unacceptable interference levels:

• mIon: between 15 A/m2 and 17 A/m2:

o greater than the threshold of 1 A/m2 for which (according to ISO 18086) the AC current density value would be indifferent.

• mIon.ac: between 35 A/m2 and 39 A/m2:

o greater than the threshold of 30 A/m2 indicated by ISO 18086 and NACE SP21424.

The explanation for this situation is given precisely by the significant level of cathodic overprotection present, represented by IR-Free potential values more negative than -1.3 V CSE and very high DC current density values, being the MP in a site suffering cathodic DC interference generated by metro and railway systems.

This results in a reduction of the Spread Resistance value, up to the point of generating an AC current density higher than the allowed limits even in the presence of a very low AC voltage.

The main evidence of the dependence of this condition on over-protection has been clearly shown when, due to a malfunction, one of the two T/Rs protecting the Cathodic Protection system did shut down, changing the values measured on the Measurement Point as in Table 2:

 

 

 

 

Combatting Corrosion, Vibration, and Fatigue Under Pipe Supports:  SmartPad System Advances for the Energy and Process Sectors

Combatting Corrosion, Vibration, and Fatigue Under Pipe Supports: SmartPad System Advances for the Energy and Process Sectors

Hani Almufti, Technical Lead, Cogbill Construction (RedLineIPS), USA.

Meet the Author

Hani Almufti is Engineer and Manager of Strategic Development at Cogbill Construction (RedLineIPS), where he leads product strategy, materials selection, and technical guidance for metallic and non-metallic pipe support systems. He holds a B.S. in Industrial Engineering and is a Master’s candidate in the same field. With 15+ years in pipe supports—including a decade focused on offshore energy corrosion—he specialises in corrosion under pipe supports (CUPS) and the performance of FRP/composite and metallic supports. His expertise spans corrosion mitigation, reliability engineering, and process improvement, with a sustained focus on reducing risk, noise/vibration, and lifecycle cost across onshore and offshore assets.

Photo 1: Installed System On Pipe Gantry.

1. Introduction

Pipe-support interfaces are convergence points for several degradation modes in industrial and offshore piping: corrosion under pipe supports (CUPS), vibration, structure-borne noise, and fatigue. On offshore platforms and FPSOs, major operators have reported that these risks are heightened by salt-laden atmospheres, hull motion, and restricted access, while the pipe–pad contact remains difficult to inspect [1].

Conventional mitigations—welded metallic pads, thermoplastic half-rounds, and epoxy-bonded plates—can re-establish galvanic paths, trap electrolyte, or require hot work and cure time. Maintaining seal integrity, alignment, and controlled slip becomes increasingly challenging as coatings wear and thermal cycles accumulate [1]. The RedLineIPS SmartPad System is a fully non-metallic support interface comprising a load-spreading FRP saddle, a bonded closed-cell elastomeric gasket (Hydroseal), and FRP bands/buckles.

Together, they electrically isolate the pipe from the steel support, seal the pipe–pad contact to discourage moisture films, provide viscoelastic damping at the interface, and relocate thermal slip to a controlled, low-friction plane on the saddle/support side. This paper outlines the design rationale, installation approach, and third-party proofs, and summarises field experience from a Gulf Coast of Mexico chemical plant retrofit programme.

2. The SmartPad System

2.1 Composite FRP SmartPad (Saddle Wear Pad)
2.1.1 Construction and geometry
Structural fibre-reinforced polymer (FRP) saddle fabricated from continuous-strand mat (CSM) in a vinyl-ester matrix, moulded to standard pipe curvatures of 1/2” to 72” NPS. The crown radius and contact width are sized to spread load over a broad arc, keep local bearing pressure low, and maintain stable seating under thermal and dynamic loads.

2.1.2 Functions at the pipe–pad interface
• Load distribution:
Spreads the pipe’s weight over a wider area so no small spot takes all the pressure—reducing dents and coating damage.

• Electrical isolation: Non-conductive composite interrupts metal-to-metal continuity (limits galvanic coupling to support steel).

• Coating protection: Smooth, inert bearing surface reduces abrasion during thermal slip and vibration.

• Offshore durability: Vinyl-ester chemistry with UV inhibitors tolerates chloride-rich, marine atmospheres.

Photo 2: SmartPad Exoskeleton with Grooves for Bands.

2.1.3 Typical Material Properties

• Resin system: Vinyl ester; reinforcement: CSM; glass content: ~35–40 wt%.

• Compressive strength (ASTM D695): ~25,000 psi
(172 MPa) [2].

• Flexural strength (ASTM D790): >30,000 psi (207 MPa).

• Continuous service temperature: -60°F to 400°F (-51°C to 204°C).

• UV resistance: High (integral inhibitors).

• Fire behaviour: Fire-retardant formulation (rating available on request).

2.1.4 Manufacture and Integration

Hand lay-up with controlled cure to achieve low void content and uniform fibre wet-out. Finished edge radii and surface roughness are controlled to minimize coating gouge. Saddle curvature and contact-width tolerances support repeatable fit and clamp preload. The FRP saddle provides the load-bearing, isolating substrate for the bonded closed-cell gasket and FRP banding within a fully non-metallic load path.

2.2 Hydroseal Closed-Cell Gasket

Photo 3: FRP Saddle and Closed Cell Gasket.

2.2.1 Construction and Placement

Factory-bonded to the SmartPad’s pipe side, the closed-cell elastomer compresses under band preload to form a continuous, conformal contact that accommodates normal surface roughness and remains uniform through thermal and vibration cycles at the pipe–pad interface.

2.2.2 Functions at the Pipe–Pad Interface

• Moisture Exclusion / CUPS Control: Very low water uptake; compressed contact suppresses crevice geometry and ion/oxygen transport, limiting crevice/under-deposit and MIC precursors.

• Vibration and Noise Attenuation: Viscoelastic damping lowers transmitted shear and micro-slip; the compliant, non-metallic layer acts as an acoustic impedance break to reduce structure-borne noise and alternating stress.

• Assists Galvanic Isolation: In combination with the FRP saddle, maintains a fully dielectric contact path.

2.2.3 Typical Material Properties

• Type: Closed-Cell Elastomer (e.g., silicone / EPDM)

• Density: 14–18 lb/ft³ (≈225–290 kg/m³)

• Compression-deflection @25% (ASTM D1056): ≈7.5 psi (≈52 kPa) [3]

• Hardness (ASTM D2240, Shore 00): 40–60

• Water absorption (ASTM D471): <0.1% by volume

• Operating temperature: –60°F to 570°F (-51°C to 300°C) • Compression-set resistance: Excellent

2.2.4 Durability and Integration

Under FRP-band preload, the gasket maintains stable compression, preserving seal and damping through thermal/vibration cycling and tolerating minor surface irregularities from prior repairs. Within the fully non-metallic load path, the gasket supplies sealing, compliance, and energy dissipation that complement the saddle’s stiffness and protect the coating at the pipe–pad interface.

2.3 SmartBands

2.3.1 Construction and Locking

Continuous long-strand FRP straps in a UV-resistant resin, paired with a matching-pitch FRP square-tooth buckle for incremental, non-backdrivable engagement. Radiused edges and smooth faces limit stress risers and coating damage.

Photo 4: Non-Metallic Straps and Buckles.

2.3.2 Functions at the Pipe–Pad/Support Interface

• Dielectric clamping: All-composite load path preserves electrical isolation (avoids galvanic reintroduction).

• Preload delivery/retention: Long-strand reinforcement improves load transfer and creep/fatigue resistance, maintaining clamp force through thermal and vibration cycling.

• Surface compatibility / constructability: Non-marring inner face; smooth outer face for clean tensioning. Installs with a calibrated handheld tool—no hot work or adhesives.

2.3.3 Typical material properties

• Material: Continuous-strand FRP; UV-resistant resin.

• Tensile capacity (per band): ~1,200 lbf (≈5.3 kN).

• Thermal range: –40 °F to 250 °F (–40 °C to 121 °C).

• Electrical behaviour: Dielectric, non-metallic.

• Environmental durability: Corrosion-immune; outdoor/UV rated for coastal/offshore service.

2.3.4 Installation and Preload Control
Bands routed in moulded circumferential grooves in the saddle engage the FRP buckle and are tensioned to specification with a calibrated tool. Groove geometry sets bend radius, keeps the strap flush/recessed, and prevents lateral migration; the low-profile routing avoids snagging and maintains uniform bearing. Preload is confirmed by tool indication (or witness marks). For underside inspection, bands are single-use—they are cut and replaced with new bands; replacements are low-cost, and reinstallation typically takes minutes per support. Grooved routing also localises relative motion to the engineered slip plane at the saddle–support interface [4].

2.3.5 Durability and Integration
The continuous-strand architecture resists creep and tooth-root fatigue under cyclic loads. UV-stabilised resin supports long outdoor exposure; the all-composite assembly is unaffected by chloride corrosion. SmartBands provide the clamping force that maintains the Hydroseal seal and the saddle’s load-sharing contact within a fully non-metallic load path.

Photo 5: Example of Corrosion Under Pipe Support.

3. Corrosion Mechanisms at Support Interfaces

3.1 Crevice / Differential Aeration

Mechanism:  A narrow, shielded gap at the pipe–pad interface traps a thin electrolyte. Oxygen is depleted inside the gap while adjacent surfaces remain aerated, creating an anode/cathode differential. Wet–dry cycling concentrates chlorides and lowers pH, undermining coatings and accelerating localised metal loss [1].

SmartPad Mitigation: A factory-bonded, closed-cell Hydroseal gasket forms a continuous conformal contact under preload, denying voids where films persist. The FRP saddle spreads load to keep contact pressure uniform through thermal cycles, disrupting the differential-aeration cell associated with CUPS.

3.2 Galvanic at the Support

• Mechanism: Electrically coupled dissimilar (or conditionally different) metals sharing an electrolyte drive anodic dissolution; small-anode/large-cathode area ratios intensify attack at supports [1].

• SmartPad Mitigation: A fully dielectric load path—FRP saddle, Hydroseal gasket, and FRP SmartBands™/buckles—breaks metal-to-metal continuity. The sealed interface also limits shared electrolyte, cutting off both prerequisites for galvanic corrosion.

3.3 Microbiologically Influenced Corrosion (MIC)

• Mechanism: In intermittently wet crevices, biofilms (e.g., SRB) create chemically distinct microenvironments (sulfides, acidity, differential aeration) that localise attack

[1]. • SmartPad Mitigation: The low-uptake, closed-cell contact shortens wet-film residence time and reduces attachment sites. Smooth, non-porous, electrically isolating surfaces further discourage biofilm establishment and persistence at the pipe–pad interface.

3.4 Fretting-Assisted Corrosion

• Mechanism: Sub-millimeter relative motion from vibration/thermal cycling abrades coatings and oxides; freshly exposed steel corrodes between slips, forming a wear–corrosion feedback loop focused at the supports [1].

• SmartPad Mitigation: Viscoelastic damping in Hydroseal stabilises the pipe–pad contact and lowers micro-slip. Required thermal movement is relocated to the low-friction saddle–support interface, while the FRP saddle’s load distribution reduces shear at the pipe wall.

3.5 Under-Deposit/Capillary Thin-Film

• Mechanism: Deposits or capillary-held films trap chloride-rich, oxygen-poor moisture that behaves like a crevice beneath the footprint [1].

• SmartPad Mitigation: The bonded, continuous interface leaves no seam for solids to wedge; closed-cell elastomer resists wicking. Moisture remains on exposed, cleanable surfaces rather than beneath the pipe.

4. Vibration

4.1 Sources and frequency content
Piping vibration originates from rotating/reciprocating equipment (pumps, compressors, blowers), pulsation in positive-displacement services, turbulence at fittings/reducers, two-phase/cavitation, hydraulic transients, alignment/soft-foot issues, and support stiffness mismatches. Field spectra commonly fall in the 10–100 Hz band with ~0.25–2.5 mm (0.01–0.10 in) peak-to-peak motion; response amplifies near span/support natural frequencies (cf. ISO 20816-1) [5].

4.2 Why the pipe–pad interface matters
Rigid, metal-to-metal load paths transmit dynamic energy as micro-slip and contact shear at the pipe–pad interface. This accelerates coating wear (promoting CUPS), excites support steel (structure- borne noise), loosens hardware, and increases alternating stress Δσ—shortening fatigue life per S–N behaviour.

4.3 SmartPad mitigation mechanisms
• Interface damping (Hydroseal).
The closed-cell elastomer provides viscoelastic damping in the 10–100 Hz range, reducing transmitted shear/micro-slip and smoothing contact pressures [6].

• Relocated slip (FRP saddle). The moulded saddle furnishes a controlled, low-friction slip plane at the saddle–support interface so thermal movement does not abrade the coating at the pipe–pad interface; broad bearing further lowers work per cycle.

• Stable dielectric clamping (SmartBands™ in recessed grooves). Calibrated, all-composite preload maintains uniform contact without re-introducing metallic short circuits, low-profile routing resists lateral migration and secondary rattles.

5. Sound (Structure-Borne Noise)

5.1 Mechanism
Dynamic forces excite the pipe wall; a rigid, metal-to-metal path at the pipe–pad interface transmits that energy into support steel and deck members, which then radiate airborne noise. Frictional micro-slip at a hard contact can also generate “stick–slip” (squeal) components. Acoustic transmissibility rises when the interface impedance closely matches the supporting structure.

5.2 Sources and Frequency Content
The same drivers as vibration—rotating/reciprocating equipment, pulsation in positive-displacement (PD) services, turbulence, two-phase/cavitation, hydraulic transients, alignment/soft-foot, and support stiffness issues—dominate. On process/offshore lines, most structure-borne content is ~20–200 Hz, overlapping habitability and communication bands

[5]. 5.3 SmartPad Noise-Control Mechanisms

• Impedance break + damping (Hydroseal): The closed-cell elastomer introduces a compliant, non-metallic layer at the pipe–pad interface, lowering mechanical impedance and adding viscoelastic loss. Result: reduced transmissibility and less friction-generated noise from micro-slip.

• Controlled slip on the support side (FRP saddle): The moulded FRP surface provides a low-friction slip plane at the saddle–support interface, keeping relative motion off the coating and suppressing stick–slip at the pipe–pad contact. Broad bearing further lowers contact forces that drive radiation.

• Dielectric, Low-Profile Clamping (SmartBands in recessed grooves): All-composite bands maintain the decoupled path
(no metallic short-circuit) and sit flush to avoid secondary rattles; calibrated preload keeps contact uniform.

6. Structural Integrity (Fatigue & Stability)

6.1 Overview
The pipe–pad interface largely governs fatigue performance at supports. A hard, rigid contact concentrates routine loads and transmits vibration into repeatable stress cycles, leading to local denting, coating loss, misalignment, and ultimately crack initiation in the pipe wall or supporting steel [7].

6.2 Principal contributors at supports
• Thermal restraint.
Limited slip forces the pipe to bear against the interface; daily temperature swings add alternating load.

• Small real contact area / edges. Narrow bearings or sharp transitions elevate local pressure and seed dents.

• Dynamic excitation. Equipment- and flow-induced vibration increases the stress range each cycle.

• Fit-up variability. Misalignment or uneven bearing amplifies local stress and accelerates coating abrasion.

6.3 Why this matters for fatigue

Fatigue life follows S–N behaviour and is controlled by the alternating stress amplitude (Δσ). Dents, coating scrapes, and other stress raisers reduce cycles to initiation; once the coating is breached, corrosion further degrades the section, compounding risk [7].

6.4 SmartPad mitigation mechanisms

• Load distribution — FRP saddle. Broad bearing lowers peak contact pressure and mitigates edge effects; the non-conductive substrate avoids metal-to-metal paths that undermine coatings.

• Compliance & damping — Hydroseal gasket. A firm, closed-cell elastomer equalises contact pressure, absorbs vibration, and cushions small impacts, reducing contact shear and Δσ per cycle [6].

• Controlled movement without abrasion — saddle–support slip plane. Thermal growth is taken on the moulded FRP surface (optional low-μ liner if needed), minimising stick–slip and fretting at the pipe–pad contact.

• Stable alignment & clamp — SmartBands in recessed grooves. Calibrated, all-composite preload keeps contact uniform and resists lateral migration; the dielectric, low-profile routing avoids galvanic short-circuits and loose hardware.
(For underside inspection, bands are single-use—cut and replaced; this is a low-cost operation).

Photo 6: Full System Assembly.

7. Third-Party Testing: SmartPad Suitability for Industrial Service

Independent third-party testing was performed on specimens, as follows:

7.1 Results – Proof Loads, No Failures Observed

• Pad-only Compression: FRP saddle on NPS 16, STD wall pipe sustained 113,300 lbf axial compression without pad failure.

• Assembly Compression: Banded SmartPad-on-pipe (4.5 in OD × ¼ in wall) sustained 26,400 lbf axial compression without pad failure.

• Assembly Shear: Same assembly sustained 1,057 lbf lateral
(shear) without pad or band failure.

7.2 Interpretation
For the geometries/fixtures tested under monotonic loading, neither the composite saddle nor the banded assembly was the limiting element. The components tolerated high localised bearing and incidental lateral restraint typical of pipe-support reactions when installed and preloaded to specification.

7.3 Scope and Limits
These are static proof tests on short specimens. They do not establish design allowables or characterise fatigue, creep/relaxation, or environmental durability. Apply normal owner engineering practices (codes, load combinations, temperature, vibration/fatigue assessment) [8,9].

7.4 Implication for Use
Combined with the corrosion mechanisms described in section 3 (sealed dielectric interface, viscoelastic damping at the pipe–pad contact, and load spreading/controlled slip), the proofs support the SmartPad System’s mechanical suitability as a non-metallic pipe-support interface for industrial and offshore service, subject to project-specific engineering review.

8. Case Study — Coastal Texas Chemical Plant (Anonymised) Background

A large Gulf Coast complex retrofitted the RedLineIPS SmartPad System to mitigate CUPS, structure-borne noise, and nuisance vibration at pipe/support interfaces in a salt-laden, high-humidity environment.

8.1 Scope

• Units: Olefins recovery, utilities/cooling water, brine handling.

• Lines: Carbon-steel piping from 2”–24” NPS; cooling-water return, light condensate, brine.

• Quantity: ~5,000 supports installed during routine windows (no hot work).

• Configuration: FRP saddle + bonded Hydroseal closed-cell gasket + FRP SmartBands/buckles.

Photo 7: Installed System at Formosa Plant.

 

Photo 8: Installed System at Formosa Plant.

 

ICorr Proudly Supports the Prestigious 72nd Hatfield Memorial Lecture in Sheffield

On the 2nd December 2025, the Institute of Corrosion (ICorr) was proud to serve as a sponsor of the prestigious 72nd Hatfield Memorial Lecture, https://sheffield.ac.uk/cmbe/school/events/72nd-hatfield-memorial-lecture hosted at the University of Sheffield. ICorr’s President, Dr Yunnan Gao, attended the event on behalf of the Institute, demonstrating ICorr’s ongoing commitment to supporting academic excellence and advancing corrosion and materials science across the UK.

A Distinguished Lecture Delivered by ICorr Fellow Professor Mary Ryan

This year’s lecture was delivered by Professor Mary Ryan, CBE, FREng, FICorr, a Fellow of ICorr and Vice Provost (Research and Enterprise) at Imperial College London, https://www.imperial.ac.uk/about/leadership-and-strategy/provost/vice-provost-research/ who is renowned globally for her work in corrosion, surface engineering, and nanomaterials. Professor Ryan offered an insightful and forward-looking presentation entitled Life and Death at the Nanoscale” that captured the attention of industry experts, academics, and students alike.

Her lecture continued the long-standing tradition of the Hatfield Memorial Lecture series, which honours metallurgist Professor William Hatfield and has become a cornerstone event for the UK materials and corrosion community.

Presentation of the Hatfield Award

During the event, Professor Mary Ryan was formally recognised for her significant academic and professional achievements. She received a commemorative award presented by Professor Joan Cordiner, Head of the Department of Chemical and Biological Engineering (CMBE) at the University of Sheffield, marking a highlight of the evening.

ICorr’s Role and Presence at the Event

As one of the official sponsors, ICorr’s support was prominently acknowledged throughout the night, including on the main auditorium screen and event materials. The presence of the ICorr President reinforced the Institute’s mission to promote collaboration, scientific excellence, and knowledge exchange across the corrosion and materials community.

ICorr’s involvement with the Hatfield Memorial Lecture reflects its dedication to fostering innovation and supporting both established and emerging experts in the field.

ICorr President Hosts CSCP President Professor Xiaogang Li for High-Level UK Visit

ICorr President Hosts CSCP President Professor Xiaogang Li for High-Level UK Visit

The President of the Institute of Corrosion (ICorr), Dr Yunnan Gao, was honoured to host Professor Xiaogang Li, President of the Chinese Society for Corrosion and Protection (CSCP) https://www.cscp.org.cn/ during a significant multi-day visit to the United Kingdom in November 2025. This most welcome visit marked another historic milestone in UK-China cooperation in corrosion science and technology and reflects the growing international influence of Professor Li – one of the world’s most distinguished leaders in corrosion science and engineering.

Professor Li, who also serves as Professor at the University of Science and Technology Beijing, Director of the National Materials Corrosion and Protection Data Centre, and Editor-in-Chief of npj Materials Degradation https://www.nature.com/npjmatdeg/ travelled to the UK with a CSCP delegation to deepen academic, industrial, and institutional collaboration with ICorr and leading UK research organisations.

Strategic Visit to The University of Manchester

The first stop of the visit was The University of Manchester https://www.materials.manchester.ac.uk/ where the ICorr President, Dr Yunnan Gao, welcomed Professor Li and the CSCP delegation. Discussions centred on establishing structured pathways for knowledge exchange, joint workshops, and collaborative programmes in corrosion science, materials degradation, and advanced characterisation.

Photo 1: Collaboration Meeting at The University of Manchester. Professor Dirk Engelberg of Manchester University Giving a Presentation.

The delegation toured Manchester’s world-class materials and corrosion research facilities, including advanced electron microscopy laboratories and the Henry Royce Institute hub https://www.royce.ac.uk/ guided by Professor Xiaorong Zhou and Professor Dirk Engelberg, both internationally recognised leaders in corrosion and structural materials.

Photo 2: CSCP President Professor Xiaogang Li (L3), ICorr President Dr Yunnan Gao (R2) and the Whole Delegation Hosted by Manchester University Professor Xiaorong Zhou (L2) and the Team at Henry Royce Institute, Manchester University.

These engagements highlighted strong synergies between UK and Chinese research capabilities in environmental degradation, data-driven modelling, and materials genome methodologies—areas strongly aligned with Professor Li’s pioneering work in corrosion big data.

Visit to the University of Leeds

The delegation then travelled to the University of Leeds https://eps.leeds.ac.uk/mechanical-engineering-research-innovation where they were hosted by Professor Richard Barker and Dr Joshua Owen. A dedicated collaboration workshop focused on shared challenges across atmospheric corrosion, mechanical performance in complex environments, and infrastructure durability.

Photo 3: Collaboration Meeting at the University of Leeds. Dr Joshua Owen Giving A Presentation.

This was followed by technical tours of the University’s corrosion laboratories and the Henry Royce Institute facilities in Leeds, where cutting-edge capabilities in electrochemical methods, advanced microscopy, and materials testing were showcased.

Photo 4: Visit to the Corrosion Laboratory at the University of Leeds.

ICorr and CSCP held productive discussions on joint collaborative initiatives, potential bilateral joint research programme, and co-developed technical guidance for emerging technologies.

Final Engagement at the National Physical Laboratory (NPL)

The final stage of the visit took place at the National Physical Laboratory (NPL) in London https://www.npl.co.uk/ hosted by Professor Gareth Hinds, a globally respected figure in electrochemistry and corrosion measurement science.

A high-level meeting explored future cooperation on standards development, corrosion testing methodologies, and the potential for coordinated international research programmes that link the UK, China, and broader global partners.

Photo 5: CSCP President Professor Xiaogang Li (L3), ICorr President Dr Yunnan Gao (Right) and China Delegation Hosted by Professor Gareth Hinds (R2) at NPL, London.

A detailed tour of NPL’s corrosion labs followed, highlighting UK leadership in precision materials characterisation, CCS (Carbon Capture and Storage) corrosion testing, and corrosion-related metrology.

Photo 6: Visit to the Corrosion Laboratory at NPL.

The visit concluded with a symbolic and impactful moment – the ICorr President and Professor Li shaking hands at the end of the visit at NPL, reaffirming the commitment of both organisations to long-term partnership.

Photo 7: ICorr President and CSCP President – Shaking Hands at the End of the Visit.

A Transformative Step for Global Corrosion Collaboration

Professor Li’s visit represents one of the most significant international engagements in ICorr’s recent history. His exceptional global standing-reflected in his pioneering contributions to corrosion big data, materials genome engineering, atmospheric corrosion mechanisms, and the development of advanced corrosion-resistant steels-amplifies the strategic importance of this collaboration.

His leadership has had profound influence across China and the international corrosion community, and his engagements in the UK underscore the strong mutual interest in scientific cooperation, talent development, and harmonisation of methodologies across continents.

ICorr Statement on the Visit

The ICorr President commented:

“This visit marks a new affirmative chapter of international cooperation. Professor Li is one of the most influential corrosion scientists in the world, and the strengthening of ties between ICorr and CSCP will accelerate innovation, unify global communities, and help tackle some of the most complex materials challenges facing society. We look forward to expanding this partnership for the benefit of both nations and the global corrosion community.”

Looking Ahead

ICorr and CSCP have agreed to continue developing:

  • Bilateral research programmes
  • Collaborative standards development
  • Exchange opportunities for students and early-career corrosion practitioners
  • Joint international webinars, workshops and conferences
  • New awards schemes
  • Shared initiatives in corrosion data science, materials design, and advanced testing

This landmark visit has not only reinforced the scientific bonds between the UK and China but has also set the stage for a new era of global cooperation in corrosion science and engineering.

 

 

 

 

Call for papers for Eurocorr 2026

Call for papers for Eurocorr 2026

The Institute of Corrosion (ICorr) is pleased to announce that the Call for Papers for EUROCORR 2026 (https://eurocorr2026.org/)- to be held from 6-10 September 2026 in Dublin, Ireland-is now open. As one of the co-organisers of this prestigious event, we warmly invite our members to contribute and participate.

The deadline for the submission of lecture and poster abstracts is 16th January 2026.

Abstracts can be submitted via the following link:

https://dechema.converia.de/frontend/index.php?sub=851

Please find ICorr’s EUROCORR 2026 Call for Papers flyer. We encourage you to share it with colleagues and within your professional networks who may be interested in contributing.

ICorr strongly encourages all members to take part in EUROCORR 2026 and help showcase the strength and innovation of our corrosion community.