Field service left the hybrid sucker rod with visible splits and measurable strength loss
Pultruded carbon/glass-fiber hybrid composite rods are built for the punishing mix of axial loading, fluid exposure, and repeated handling that defines rod-pumping service. This PLOS ONE study tracked what happened after one such rod returned from the field with a longitudinal split that developed during retrieval and drum winding after 435 days of actual downhole use. The failure was not treated as an isolated break. It was examined as the visible end point of a longer degradation process that likely accumulated during service and then advanced when the rod was bent again during recovery operations.
Five specimens cut from a pristine 22 mm parent rod were compared with five specimens cut from a service-aged parent rod. The design allowed direct comparison between unused material and field-exposed material across tensile, compressive, flexural, and apparent horizontal shear testing, alongside scanning electron microscopy, dynamic mechanical analysis, moisture-desorption measurements, and bootstrap-assisted Weibull analysis. The result was a detailed picture of how mechanical deterioration and interfacial damage unfolded across the rod’s layered architecture, including the outer glass-fiber region, the carbon/glass hybrid transition region, and the inner carbon-fiber core.
The most consequential changes appeared in bending-sensitive and interface-sensitive properties. Mean tensile strength fell by 10.02%, mean flexural strength by 19.92%, and mean apparent short-beam shear strength by 21.98% after service. After Holm correction, the flexural and apparent short-beam shear reductions remained statistically significant, while the tensile-strength decline did not reach significance with adjusted p = 0.057.
Bending and shear weakened more than pure axial strength
The pattern of degradation fits the mechanics of a rod that spends most of its life under alternating tension and compression, then faces extra bending when it is retrieved and wound on a drum. Axial tensile loading can damage the composite gradually, but flexure and short-beam shear are more sensitive to matrix cracking, debonding, and crack growth along interfaces. In a hybrid rod, the transition zone between carbon and glass reinforcement is a natural stress-transfer region, so any loss of matrix continuity there can quickly reduce the ability to carry off-axis load.
Compressive strength changed far less. The mean decline was only 2.25%, and the difference was not statistically significant, yet the variability changed sharply. The coefficient of variation rose from 2.42% to 18.88%, while the Weibull shape parameter estimate dropped from 41.31 to 5.27. In practical terms, the material did not simply get weaker on average; it became less predictable. A rod with wider strength dispersion is harder to trust under field loading because localized weak zones can fail well before the nominal average would suggest.
The hybrid transition region showed the clearest structural damage
Microscopy gave the mechanical findings a visible substrate. Scanning electron microscopy revealed extensive matrix fragmentation, resin loss, fiber–matrix debonding, and longitudinal interfacial cracking. These features are exactly what would be expected when cyclic loading and fluid exposure act together over time. The matrix carries shear and keeps fibers locked in place; once it fragments, stress redistributes unevenly, and the interface begins to fail in strips rather than as a continuous load-bearing bridge.
The carbon-fiber core also changed, but in a more complex way. Its storage modulus decreased by 34.3%, while the glass-transition temperature increased from 159.5 to 164.76 °C. Those shifts may reflect competing effects from residual post-curing and moisture-related influences, although the underlying chemistry was not directly measured. The study also found greater equilibrium mass loss during drying in the service-aged specimens, consistent with greater retention of removable moisture. Water inside a composite can plasticize the matrix, weaken the interface, and alter thermal behavior, yet post-curing can move the opposite way by increasing chain mobility constraints. The two effects can coexist, leaving thermal and mechanical markers that do not point in a single direction.
Strength dispersion rose as the rod aged in the wellbore
The broad statistical tools used here matter because service damage does not distribute evenly. One section may remain relatively intact while another accumulates matrix cracks, interfacial separations, and local fiber exposure. Welch’s t-tests compared mean property changes, but the Weibull analysis exposed another layer of risk: the aged rod became more heterogeneous. Lower shape-parameter estimates indicate a broader spread in failure strength, which is the sort of variability that can decide whether a rod survives a difficult retrieval or splits along its length.
The longitudinal splitting observed during recovery and winding is consistent with that risk profile. Once cracks and debonded interfaces already exist from downhole exposure, added bending can drive those defects to link up. The rod then fails not as a pristine member under a single overload, but as a damaged structure whose defects have merged into a continuous split. In that sense, the visible fracture reflects two linked phases: slow damage accumulation in service, then crack propagation during handling.
Two phases of damage explain the field failure pattern
The study’s central interpretation is a two-stage damage-evolution process. During prolonged downhole service, repeated axial loading and fluid exposure appear to have weakened the matrix and hybrid interfaces. During retrieval and drum winding, additional bending likely accelerated the growth and coalescence of existing cracks, producing the local longitudinal split seen in the field. That sequence is mechanically plausible for a three-region hybrid rod because the transition zone concentrates stress transfer and is especially vulnerable to matrix loss and interfacial debonding.
The work is also limited by what it did not measure directly. The underlying chemical mechanisms behind the thermal shifts were not characterized, and the study relied on a small number of specimens from one service-returned parent rod and one pristine comparison rod. Even so, the combination of mechanical testing, moisture-desorption data, thermal analysis, and microscopy offers a coherent failure map. It shows that service aging in this material is not confined to a single property or a single layer. It moves through the rod architecture unevenly, and when the weakened structure is bent again outside the wellbore, the accumulated damage can convert into a visible split.