Retraction: Study Examining Graphene Oxide/Nanozirconia Effects on Geopolymer Concrete Strength and Durability

A study that attracted attention for its proposed route to stronger, more durable and potentially lower-impact concrete has been retracted after editors identified apparent overlaps among scanning electron microscopy images and detected repetitive patterns in an energy-dispersive X-ray spectrum. The retraction concerns “Experimental and theoretical investigation of the influence of graphene oxide/nanozirconia on the mechanical and durability properties of geopolymer concrete,” published in Polymer Bulletin. The journal’s editors said that they could no longer have confidence in the data because the authors did not provide the underlying raw material requested during the investigation. The authors also did not respond to correspondence from the editor or publisher about the retraction.

The original article, published on 6 December 2024, examined a class of cement-free or cement-reduced construction materials known as geopolymer concrete. Conventional Portland cement production releases substantial quantities of carbon dioxide because limestone must be heated to high temperatures and chemically decomposed. Geopolymers instead use aluminosilicate-rich materials, such as industrial by-products or other mineral sources, activated with alkaline solutions. During curing, dissolved silicon and aluminium species reorganize into a three-dimensional aluminosilicate network. That network can bind aggregates and develop mechanical strength, although its performance depends strongly on precursor chemistry, activator concentration, water content, curing conditions and the microstructure formed during reaction.

The study focused on two nanoscale additives: graphene oxide and nanozirconia. Graphene oxide consists of atomically thin carbon sheets decorated with oxygen-containing chemical groups. Those groups can improve dispersion in water-based mixtures and provide sites for interaction with the geopolymer gel. In principle, well-dispersed graphene oxide could bridge microscopic cracks, refine pores and increase resistance to mechanical damage. Nanozirconia, composed of extremely small particles of zirconium dioxide, is chemically stable and mechanically hard. Added to a cementitious or geopolymeric matrix, it might act as a reinforcing filler, occupy voids and alter the interface between aggregates and the binder. Such mechanisms are plausible, but they must be demonstrated through reproducible testing rather than inferred from attractive images or isolated strength measurements.

Microscopy was central to the paper’s evidence. Scanning electron microscopy, or SEM, produces high-magnification images by scanning a focused electron beam across a specimen and recording signals generated from the interaction between electrons and the material. Depending on the detector, the resulting image can reveal surface texture, cracks, pores, particles and the morphology of reaction products. For geopolymer research, SEM images are often used to support claims about a dense binder, improved particle packing or the formation of a more continuous gel. Yet SEM images are not automatically unique fingerprints of a sample’s behaviour. Magnification, contrast, cropping, rotation and image processing can all affect how a structure appears, which is why researchers must retain raw files, document acquisition conditions and make comparisons across independently prepared specimens.

The editors reported three specific concerns about the figures. Figure 1a appeared to overlap with Figure 11 of a separate cited work, while Figure 11d appeared to overlap with Figure 3d of another cited work. The retraction notice also states that Figures 13d and 14b appeared to overlap after rotation. These observations do not merely involve images that look generally similar because they depict comparable materials; the notice describes apparent overlap in particular figures, including an instance in which rotation was involved. In a materials-science paper, a duplicated or reused micrograph can misrepresent the morphology of a different specimen, treatment or test condition. That can undermine the chain of evidence connecting a formulation to a claimed improvement in strength, durability or chemical performance.

The notice raised a second issue involving energy-dispersive X-ray spectroscopy, or EDX. EDX is commonly attached to an SEM and measures characteristic X-rays emitted when the electron beam excites atoms in a sample. Because each element produces a distinctive set of X-ray energies, the technique can help identify the elements present and estimate their relative abundance. In geopolymer studies, EDX may be used to examine distributions of silicon, aluminium, oxygen, zirconium or other elements and to support interpretations of reaction products or additive incorporation. The editors said that the background noise in the EDX plots in Figure 1 showed repetitive patterns. Background noise is expected in spectroscopy, but suspiciously repeated structures can raise questions about whether a signal was independently measured, copied, processed or generated through an inappropriate workflow.

The absence of raw data made those concerns impossible for the editors to resolve. Raw SEM and EDX files can contain information that is not visible in a published figure, including acquisition parameters, detector settings, scale calibration, sample identifiers and the unprocessed signal. Investigators can compare those files with the displayed panels, inspect whether an image has been rotated or reused, and determine whether spectral features arise from the specimen or from data handling. Without the underlying records, an editor may be unable to distinguish an honest figure-preparation error from a more serious problem affecting the reliability of the results. The retraction notice does not assign a specific cause for the apparent overlaps or repetitive patterns; it states instead that the unresolved concerns led the editors to withdraw confidence in the presented data.

That distinction matters because the original paper’s subject sits at the intersection of nanomaterials engineering and infrastructure research, fields in which experimental claims can influence subsequent formulations and testing programs. If graphene oxide or nanozirconia appears to improve compressive strength, crack resistance, water absorption or chemical durability, later researchers may use those reported proportions as starting points. Engineers may also cite microstructural evidence when assessing whether a material can withstand freeze-thaw cycles, aggressive chemicals, moisture movement or long-term loading. A compromised image does not automatically prove that every mechanical measurement is wrong, but it weakens the support for the interpretation and makes it difficult to know which conclusions, if any, remain dependable.

Geopolymer concrete research is particularly sensitive to microstructural interpretation because its properties emerge from several overlapping scales. At the molecular and nanometre scales, alkaline activation dissolves portions of the precursor and forms binding gels. At larger scales, unreacted particles, pores, interfaces and cracks govern transport and failure. Water can move through connected pores, carrying dissolved ions that accelerate degradation or trigger further reactions. Nanoparticles may alter nucleation, packing and gel connectivity, but their effects depend on dispersion. Graphene oxide can restack into sheets if poorly mixed, while nanozirconia can agglomerate into clusters that create defects rather than reinforcement. A credible claim therefore requires more than a visually dense SEM field: it calls for carefully controlled mixtures, replicated specimens, transparent mechanical and durability data, and analytical results that can be independently checked.

The retraction does not establish that graphene oxide, nanozirconia or geopolymer concrete cannot be useful. Instead, it removes one published study as a reliable basis for judging the particular experimental and theoretical claims it presented. The episode highlights why data stewardship is as important as novelty in fast-moving materials research. Researchers need to preserve original microscopy files, complete spectra, laboratory logs, specimen histories and statistical records, while journals and institutions need procedures that allow questionable images to be examined efficiently. Independent replication remains essential, especially when a proposed additive is promoted as a way to improve both performance and sustainability. For readers, the most consequential result of the notice is not a verdict on nanomodified concrete as a technology, but a warning that promising engineering narratives must rest on evidence that remains traceable from raw measurement to published conclusion.