Graphene and Graphene Oxide Look Identical — Zebrafish Behavior Says Otherwise
Same element, same 2D sheet structure — wildly different real-world behavior, in the fish that swim in it.
Graphene (GR) and its oxidized cousin graphene oxide (GO) are both single- or few-layer carbon sheets, and both are increasingly used in biomedical, sensing, and drug-delivery applications. But GO’s oxygen-functionalized surface makes it hydrophilic and water-dispersible, while GR stays hydrophobic — a structural difference that could plausibly translate into very different environmental behavior once these materials end up in waterways.
The Obstacle — Toxicity Testing Lagged Behind Adoption, and the Doses Weren’t Realistic
Most prior GR and GO toxicity studies used concentrations far above what’s actually predicted to occur in natural water systems (roughly 0.001–1 mg/L), and few examined behavioral effects in adult zebrafish specifically — despite behavioral endpoints being established as more sensitive and cost-effective toxicology indicators than many biochemical assays alone. What was missing was a study that tested both materials, side by side, at genuinely environmentally relevant concentrations, across a full battery of behaviors plus the underlying brain chemistry.
The Innovation — A Full Behavioral and Biochemical Battery at Realistic Doses
In a study published in Environmental Pollution (Audira, Lee, Siregar, Malhotra, Rolden, Huang, Chen, Hsu, Hsu, Ger & Hsiao, 2021, Environ. Pollut., 278, 116907, DOI: 10.1016/j.envpol.2021.116907), researchers at Chung Yuan Christian University, National Pingtung University, and the University of Santo Tomas chronically exposed adult zebrafish to GR (0.1 and 0.5 ppm) or GO (0.1 and 1 ppm) for about 14 days, then ran a five-test behavioral battery — novel tank exploration, mirror biting, predator avoidance, conspecific social interaction, and shoaling — followed by ELISA-based brain biochemistry measuring oxidative stress, cortisol, and a panel of neurotransmitters.
Before any of that biological testing began, the team fully characterized both materials — Raman spectroscopy, XRD, TEM, AFM, UV-vis, and zeta potential — to establish exactly what structural and physical differences distinguish the GR and GO batches actually used in the exposures.
Key Results — Two Materials, Two Behavioral Signatures, One Shared Chemistry Story
The case builds across four results: quantify the structural differences between GR and GO, show how each alters locomotion and exploration, show how each alters higher-order social and predator responses, and connect both back to brain chemistry.
Raman Spectroscopy Quantifies How Much More Defective Graphene Is Than Graphene Oxide
Raman spectra, collected on a ProTrusTech RAMaker microscopic Raman system (532 nm), showed both GR and GO with characteristic D and G bands near 1350 and 1580 cm-1. The D-to-G band intensity ratio (ID/IG) — a standard proxy for structural defect density — came in at 1.2 for GO versus a notably higher 1.9 for GR, indicating more chemical and structural defects in the reduction-processed GR sample. Combined with XRD (GO’s characteristic (001) peak absent in GR), TEM/AFM sizing (GR sheets ~5.4 µm, GO ~3.0 µm; both 2–4 atomic layers thick), and zeta potential (both highly stable, below −80 mV), this panel establishes precisely what “graphene” and “graphene oxide” mean as physical objects in this study.
Low-Concentration GR and High-Concentration GO Both Suppress Zebrafish Locomotion
In the novel tank test, both a low dose of GR (0.1 ppm) and a high dose of GO (1 ppm) significantly decreased average speed and rapid-movement time ratio, with the 1 ppm GO group showing the most pronounced hypoactivity of any treatment. Interestingly, the higher GR dose (0.5 ppm) trended toward hyperactivity instead — a non-monotonic pattern that doesn’t simply track “more material, more suppression.” Both GR groups (but neither GO group) also showed reduced thigmotaxis, an altered exploratory signature specific to graphene.
Both Materials Blunt Predator Avoidance — GR at the Higher Dose Most of All
Neither material significantly altered mirror-biting aggressiveness, but predator avoidance told a different story: chronically GR- and GO-treated zebrafish spent significantly more time approaching a live predator and kept a significantly shorter average distance from it, with the 0.5 ppm GR group showing the most severe alteration. Since reduced predator avoidance directly threatens survival in the wild, this endpoint carries more ecological weight than the locomotor changes alone.
0.5 ppm GR Suppresses Neurotransmitters and Cortisol; Both Materials Elevate Oxidative Stress
ELISA analysis of zebrafish brain tissue found that 0.5 ppm GR significantly decreased serotonin, acetylcholine, and dopamine levels, along with cortisol — a pattern consistent with the behavioral changes seen at that dose. Meanwhile, reactive oxygen species (ROS), a marker of oxidative stress, was significantly elevated in both the 0.1 ppm GR and 1 ppm GO groups — the two doses that, not coincidentally, also produced the most pronounced locomotor and predator-avoidance changes for their respective materials.
Why ProTrusTech Is Featuring This Research
Every downstream behavioral and biochemical claim in this paper depends on knowing precisely what GR and GO are, structurally, in the specific batches tested — and that baseline was established with Raman spectroscopy on a ProTrusTech RAMaker system. The D/G band ratio isn’t a footnote: it’s the number that lets the authors say, with confidence, that their GR sample carries meaningfully more structural defects than their GO sample — relevant context for why two chemically similar materials produce different, sometimes opposite, behavioral signatures.
Engineering the Solution
The characterization workflow — Raman for defect density, XRD for layer count and crystallinity, TEM/AFM for size and thickness, UV-vis for electronic structure, and zeta potential for colloidal stability — gives a complete physical fingerprint of a 2D nanomaterial batch using standard, non-destructive techniques, all before a single fish is exposed. That fingerprint is what makes the behavioral results interpretable rather than merely observational.
Featured Instruments:
Why This Matters
This paper is part of a cluster of zebrafish nanomaterial toxicology studies from an overlapping author team at Chung Yuan Christian University and National Pingtung University — alongside a copper oxide nanoparticle / carbofuran cardiotoxicity study and a multi-walled carbon nanotube toxicity study, both also featured here and both also using RAMaker Raman spectroscopy to verify material identity before biological testing. Read as a set, the three papers make a consistent methodological case: whether the endpoint is cardiac rhythm, brain neurochemistry, or adult behavior, and whether the material is a metal-oxide nanoparticle, a carbon nanotube, or a 2D carbon sheet, this group treats Raman-confirmed material characterization as a non-negotiable first step before any zebrafish enters an exposure tank.
What’s Next
- Longer exposure windows and lower, more chronic concentrations. The current 14-day exposure at 0.1–1 ppm already sits at the low end of environmentally predicted concentrations; extending duration would test whether effects intensify, plateau, or resolve.
- Mechanistic follow-up on the GR defect-density / behavior link. The paper establishes a correlation between structural defect density (via Raman) and behavioral outcome; a direct causal test — high- vs low-defect GR head-to-head — remains open.
- Extending the battery to F1 offspring or developmental exposure. This study used adult zebrafish; whether chronic parental or direct embryonic exposure compounds the effects is untested here.
Working on 2D nanomaterial characterization, zebrafish behavioral toxicology, or Raman-based defect-density quantification of your own? Get in touch with our team.
Key Takeaways
- Raman spectroscopy on a ProTrusTech RAMaker system quantified graphene’s defect density (ID/IG = 1.9) as notably higher than graphene oxide’s (1.2), establishing the material baseline before any behavioral testing.
- At predicted environmental concentrations (0.1–1 ppm), both GR and GO altered zebrafish locomotion, with direction and magnitude varying non-monotonically by dose and material.
- Both materials significantly blunted predator-avoidance behavior, an ecologically consequential endpoint, with 0.5 ppm GR the most severe.
- 0.5 ppm GR decreased brain serotonin, acetylcholine, dopamine, and cortisol, while both 0.1 ppm GR and 1 ppm GO elevated oxidative stress (ROS).
- This completes a three-paper cluster of RAMaker-verified zebrafish nanotoxicology studies from the same overlapping group, spanning cardiac, neurochemical, and behavioral endpoints.
Featured Publications:
Related publications from the same author teams: Saputra, F. et al., “Co-Treatment of Copper Oxide Nanoparticle and Carbofuran Enhances Cardiotoxicity in Zebrafish Embryos,” featured here as Copper Oxide Nanoparticles and a Pesticide Synergize to Disrupt a Zebrafish Heart; and Audira, G. et al., “Assessments of carbon nanotubes toxicities in zebrafish larvae,” Chemico-Biological Interactions, 2024, 392, 110925, featured here in our carbon-nanotube zebrafish study.
