Parametric 3D finite element analysis of FRCM-confined RC columns under eccentric loading

https://doi.org/10.1016/j.engstruct.2020.110504Get rights and content

Highlights

  • A detailed 3D FE model for slender FRCM-confined RC columns.

  • Studying cross-section shape (circle, square, hexagon, and octagon).

  • Studying aspect ratio (h/b) and slenderness ratio (KL/r) for rectangular columns, varying (h/b) from 0.5 to 2.5, and (KL/r) from 10 to 75.

  • Studying eccentricity (e) as ratio of the side (h), varying (e/h) from 0 to 2.5.

  • A noticeable change of behavior at e/h = 0.5, possibly due to shift from axial to bending.

Abstract

Fiber reinforced cementitious matrix (FRCM) is emerging as a viable retrofit and confinement technique, in lieu of fiber reinforced polymer (FRP) system which suffers from a number of issues related to the use of synthetic binders. While many studies have been conducted on the use of FRCM in shear and flexural applications, few were dedicated to confinement of slender columns, particularly those related to finite element (FE) analysis. In this study, a nonlinear three-dimensional FE model has been developed to study the behavior of reinforced concrete (RC) columns confined by (FRCM) jackets, and loaded concentrically and eccentrically. Drucker-Prager (DP) concrete model, which has several improvements over traditional DP models, was used to model the concrete core. Composite failure in the fibers comprising FRCM system and column buckling were also considered in the developed FE model. The model was validated by comparing its predictions with those of three control and 8 FRCM-confined RC columns from literature. Consequently, a parametric study utilizing 96 additional models, was performed on five parameters, namely: cross-sectional shape (square, circle, hexagon, and octagon), and for rectangular columns; aspect (h/b) ranging from 0.5 to 3, at 0.5 increment; slenderness (KL/r) ratio, considering four values, 10, 25, 50, and 75; load eccentricity (e) as a ratio (e/h) to side length (h), varying from 0 to 2.5; and concrete compressive strength (c), studying three values: 20, 35, and 50 MPa. Effects of these parameters on the column’s maximum load (Pmax) and general behavior, are discussed in details in Section 6 and summarised in the conclusions part. In general, Pmax increased by 0–32% as a result of applying one layer of FRCM jacket, and showed great dependence on the examined parameters.

Introduction

Reinforced concrete structures are facing a major deterioration problem globally due to aging; exposure to environmental conditions such as freeze-thaw cycles, moisture, de-icing agents; lack of maintenance; etc. [1], [2]. The use of fiber reinforced polymer (FRP) composites has gained considerable popularity due to their favorable properties, including low weight-to-strength ratio, ease of application, corrosion resistance, and minimal change in structure’s geometry [3], [4], [5]. In concrete retrofit applications, FRP reinforcement is typically applied as an externally bonded (EB) or as near surface mounted (NSM), both are installed using organic adhesives such as epoxy [1], [7]. Epoxy-bonded FRP strengthening techniques have some drawbacks, including: inapplicability on wet surface or at low temperature, high cost, poor performance under high temperature and lack of vapor permeability [1], [7].

In order to overcome these issues, researchers examined the replacement of epoxy resins with inorganic (cement mortar) matrix [1], [6], [7], [8], [9], [10], [11]. The mortar can be used to bond EB FRP and NSM FRP reinforcements, but it’s generally more effective with fabric meshes and grids — resulting in a retrofit system known in literature as fiber-reinforced cementitious matrix (FRCM) [7]. Different types of fibers, such as carbon, glass, basalt, or Polyparaphenylene Benzobisoxazole (PBO), can be used in fabricating the FRP mesh/grid [10] in FRCM technique. Several tests have been performed to validate the effectiveness of FRCM system in strengthening concrete and masonry members against shear, flexure, and torsion, with results showing significant increase in load capacity and reduction in deflections and crack widths, in addition to improved behavior under elevated temperatures or in fire [6], [7], [9], [10], [11], [12], [13], [14].

Research on strengthening or confinement of axial RC members using FRCM technique can also be found in literature, although in smaller numbers than other applications [15], [16], [17], [18], [19], [20]. Available studies can be divided into two main categories: short (plain) concrete columns such as [15], [16], [17] and slender reinforced concrete columns [18], [19], [20]. Variables investigated ranged from material related ones (concrete strength, fiber type), geometrical related ones (length and cross-section), and eccentricity of axial load. It was found that the FRCM jacketing had a significant contribution to improving strength and deformability of the tested samples.

However, research on FRCM-confined slender RC columns is very limited. To the authors’ best knowledge, the studies by Trapko [18], [19] and by Ombres and Verre [20] were the only ones found in the literature on this topic. Trapko [18], [19] tested 15, 1500 mm-long square RC columns under concentric and eccentric axial load. Three specimens were un-strengthened, while the remaining 12 were strengthened with FCRM jackets having different number of layers and orientations. Two eccentricities (e) were examined, namely 16 and 32 mm, corresponding to an eccentricity-to-height (h) (e/h) ratio of 0.080 and 0.16, respectively. Results showed an increase of ultimate load (Pult.) by 13.5, 17.7, and 4.7% relative to the control columns, when one transverse layer, two transverse layers, and one transverse combined with one longitudinal layers of FRCM were used, respectively. However, Pult. decreased in average by 11 and 31%, when the eccentricity increased from 0 to 16 and 32 mm, respectively.

Ombres and Verre [20] tested eight slender RC columns, confined by FRCM technique. The variables examined were: eccentricity(e/h) ratio from 0 to 0.33; and number of FRCM layers, varying from one to two. The results were similar to those observed by Trapko [18], [19] where the increase in ultimate load for FRCM-strengthened columns varied between 20% and 39%. In addition, it was found that the strength gain was inversely proportional to the eccentricity value. Given the small number of studies and limited parameters examined, further research on FRCM-confined RC columns is warranted. In addition, there is no finite element (FE) study performed on this research topic.

Validated FE models can be powerful analytical tool, where it typically results in tremendous reductions of time and cost, compared to experimental tests. Several studies utilized FE analysis in studying the behavior of RC members strengthened with FRCM, focusing on flexural, torsional, and bond aspects [21], [22], [23], [24]. For example, Alabdulhady et al. [21] developed an LS DYNA FE model to study the behavior of FRCM-strengthened RC beams under torsion and reported a good match between experimental and numerical results. Kadhim et al. [24] used ANSYS software to develop a FE model, studying the response of RC beams strengthened in shear with FRCM wraps and concluded of good calibration between numerical and test data.

Section snippets

Significance and objectives

In many instances, axial concrete members sustain damage due to impact or seismic activity and therefore require rehabilitation and upgrade, conventionally by FRP sheets and wraps. FRCM technique provides the advantages of conventional FRP systems, while minimizing the problems related to organic adhesive. The effectiveness of FRCM technique in confining or strengthening RC columns, especially slender ones, has not been fully examined and understood. The objective of this paper is to develop

Model description

The numerical models for the columns in [18], [19] were developed using the commercial finite element software ANSYS APDL 17.2 [26]. Due to the un-symmetric eccentric loads present in some specimens and to provide a uniform modeling procedure, a full-size column model was constructed for all specimens. Fig. 1(c) shows various parts of the FE model. A mesh sensitivity analysis was conducted, which showed that an element side of 25 mm for the concrete core and other parts, is a suitable balance

Model validation

The model described in the previous section is used to simulate the eccentrically loaded RC columns confined with FRCM technique, tested by Trapko [18], [19]. Table 1 describes the geometry and confinement layout of tested columns, and summarizes the key results from the FE models and experimental tests. The main comparison in this table is the ultimate load of RC columns. It can be noticed from this table that the ultimate load values obtained from FE models are predicted well with the maximum

Parametric study

The validated model was used in a comprehensive parametric study, comprising 96 new models, to examine the effects of key variables expected to impact the behavior of FRCM-confined RC columns. These parameters were the column cross-sectional shape; and for rectangular columns, the aspect ratio (h/b); slenderness ratio (KL/r); eccentricity (e) with emphasis on (e) outside of the column section; and concrete compressive strength (c). For each parameter, both un-strengthened and FRCM-confined

Conclusions

In this study, a detailed three-dimensional finite element (FE) model was developed to study the behavior of RC columns confined by fiber-reinforced cementitious mortar (FRCM), under concentric and eccentric loads. The model utilized a new, improved version of the Druker Prager (DP) model specifically designed for concrete axial members, with a capability of including nonlinear compression, and softening/hardening in both tension and compression. The failure in fiber mesh and buckling in the

CRediT authorship contribution statement

Akram Jawdhari: Conceptualization, Methodology, Project administration, Resources, Software, Supervision, Validation, Writing - original draft, Writing. Ali Hadi Adheem: Conceptualization, Data curation, Formal analysis, Investigation, Software, Validation, Visualization, Writing - original draft. Majid M.A. Kadhim: Conceptualization, Methodology, Resources, Software, Supervision, Visualization, Writing - original draft, Writing - review & editing.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

References (38)

Cited by (29)

  • Sandwich concrete beams with GFRP side faces tested and modeled in flexure

    2022, Thin-Walled Structures
    Citation Excerpt :

    Due to symmetry and to minimize computational efforts, only quarter-beam models were generated, Fig. 12. The key features implemented in the model are: (a) using William and Warnke [27] five-parameter material model to simulate concrete nonlinearity, cracking and crushing failures; (b) using Hashin damage criteria [28] to predict FRP failure whether by tensile or compressive ruptures, shearing, or delamination; and (c) using contact-target elements in conjunction with a Coulomb friction law [29] to simulate slipping of SIP form from concrete. The concrete, SIP forms, and steel plates added at loading and support locations were meshed with 8-node brick elements, using a mesh size with a maximum side length of 20 mm for the concrete core and steel plates, and 10 mm for the SIP form.

View all citing articles on Scopus
View full text