When Two Carnelians Look Alike but Behave Differently: Diffusion, Defects, and the Limits of Visible Evidence

Two Stones, One Name, Different Physics

Two carnelian cabochons can leave a gemologist with opposite impressions. One is consistently orange-red throughout, and a scratch made on its base—where it does not show—reveals the same color inside. The other is intensely saturated on its polished face but noticeably paler near a chip on the pavilion, and its color fades as the stone is ground down. Visually, in a mount, the two may be nearly interchangeable. Physically, they are not the same object at all.

Carnelian is the orange-red to brownish-red variety of chalcedony, a microcrystalline or cryptocrystalline form of silica (SiO₂). Its color is usually attributed to iron-bearing species incorporated during or after silica deposition. That single sentence hides the real problem: the same apparent color can arise from iron distributed through the entire chalcedony mass, from iron concentrated near the surface, or from a color center produced by irradiation. Those are different physical states, and they respond differently to heat, light, fracture, and chemical exposure. Distinguishing them is one of the more instructive problems in treated-gem science.

Why Chalcedony Is Not a Simple Crystal

Quartz is a framework silicate built from SiO₄ tetrahedra sharing all four oxygens, producing a three-dimensional network with the symmetry of the trigonal crystal system. A macroscopic quartz crystal has long-range translational order. Chalcedony is different in scale, not in fundamental chemistry. It consists of fine fibrous or granular quartz-like domains, with domains typically on the order of tens to hundreds of nanometers across, and it contains variable amounts of water and structural imperfections.

That microstructure matters for two reasons. First, the enormous internal surface area along domain boundaries and micropores provides pathways for fluids and dissolved ions to enter and react. Second, the abundant boundaries and defects create sites where impurity ions can sit in positions and oxidation states that would be less favorable in a well-ordered quartz lattice. It is precisely this openness that makes chalcedony a candidate for both natural color development and artificial modification.

Three Competing Explanations for Red Carnelian

Iron substitution and iron oxide inclusions

The most widely accepted explanation for natural carnelian color involves iron. In the precursor material—often grayish or pale chalcedony—iron is present largely as Fe³⁺, but the specific speciation and distribution determine whether the stone appears red. Heating or natural weathering can promote the formation of finely dispersed iron oxide species, including hematite-like phases, within the silica matrix. These particles are small enough to remain below the resolution of ordinary optical microscopy and act as a pigment dispersed through the bulk material.

When iron oxide is dispersed throughout the mass, the color is body color: it belongs to the material itself. Cutting, fracturing, or wearing the stone does not remove it, because there is no reservoir of color at the surface to exhaust.

Surface and near-surface diffusion

Diffusion treatment is a distinct process. In outline, the material is exposed to a source of a coloring element, often iron, at elevated temperature, and that element migrates into the outer portion of the stone. The result is a compositional gradient: a relatively iron-enriched rim overlying a less modified core. The color is produced by the same general iron-oxide chemistry as above, but its spatial distribution is imposed from outside rather than inherited from the original silica deposition.

Diffusion is not the same as heating alone. Heating alters existing chromophores, oxidation states, or mineral inclusions within a fixed bulk chemistry. Diffusion introduces additional chromophore from an external source and drives it inward. The two treatments may be combined, and their effects can overlap, which is one reason treatment attribution is often probabilistic rather than definitive.

Irradiation-induced color centers

A third mechanism involves crystal defects rather than iron. Ionizing radiation can displace atoms or trap electrons and holes at pre-existing defect sites, creating color centers that absorb visible light. Certain smoky and amber-colored varieties of quartz and chalcedony are conventionally explained in this way. The resulting color depends on the nature and concentration of the defects and on the presence of charge-compensating impurities such as aluminum.

The diagnostic consequence is that irradiation-induced color is a property of defect populations. Heating to modest temperatures can anneal some of these centers and bleach the color, and the effect is not necessarily accompanied by a detectable change in bulk iron content. Irradiation, bulk iron, and iron-oxide pigmentation therefore leave different physical fingerprints.

What Each Behavior Actually Reveals

The most important practical insight is that a stone's physical behavior is a record of how its color is distributed and what produces it. Several behaviors are worth separating carefully, because they are frequently conflated.

• Uniformity through the section. Color that persists into the interior when the stone is examined at a fracture or an unpolished base is consistent with bulk pigmentation. Color confined to a rim is consistent with a diffusion profile or a surface treatment.
• Temperature response. Iron-oxide coloration is generally stable at temperatures far below the melting point of silica. Some irradiation-induced color centers are less stable and may be partially removed by heating. Observing reversible or partially reversible color loss is suggestive but not by itself proof of irradiation, because multiple defect types and impurities can influence thermal stability.
• Chemical etching or fracturing. If a coloring layer is a coating or a shallow diffusion zone, mechanically or chemically removing that zone changes the appearance. A coating is a distinct case: it alters surface reflection and can be removed by abrasion or solvents, while diffusion has modified the lattice or near-surface chemistry and cannot be scrubbed off.

How Labs Distinguish the Mechanisms

No single instrument answers the whole question, and the following is a description of general analytical logic rather than a universal protocol.

Microscopy and microscopic sectioning can reveal color zoning, a color-banded rim, and textural inhomogeneity. However, a visual color band does not by itself prove diffusion treatment. Natural chalcedony can develop banded and zoned coloration during deposition and later weathering, producing a superficially similar pattern.

Trace-element analysis can measure iron concentration as a function of position when depth profiling is feasible. A pronounced near-surface enrichment is consistent with diffusion. The complication is that natural weathering and the original depositional environment can also enrich iron near surfaces and fractures, so a gradient alone is not a unique signature.

Absorption spectroscopy probes how the material absorbs visible and near-infrared light. Different iron species—Fe³⁺ in different local environments, iron oxide nanoparticle phases, and defect centers in quartz—produce different absorption behavior. Spectroscopy can therefore inform which chromophore dominates, but interpreting the spectra requires reference libraries and awareness that overlapping features from several species can occur in the same stone.

Raman spectroscopy probes lattice vibrations and can help identify silica phases and certain mineral inclusions. It is not a general-purpose treatment detector, and a good match to a quartz or moganite reference does not resolve whether the color was diffused, because Raman scattering in this context is not primarily reporting on trace-iron distribution.

Thermal treatment experiments as an analytical tool can show whether color changes on heating, but applying heat to a client's stone is not a routine non-destructive test. Where heating is used at all, it is done on reference material or with the understanding that the test may alter the object. This is an important limitation: some of the most direct experiments on treatment stability cannot ethically be applied to unknown finished gems.

Why the Same Color Can Be a Different Physical State

The broader scientific lesson is that color is an interaction among illumination, absorption, and the electronic and structural environment of the chromophore—not a simple label that names the material. Carnelian is an unusually clear illustration because the same visual outcome can be produced by:

• iron oxide pigment dispersed throughout the silica matrix,
• a diffused iron-enriched zone confined to the outer portion,
• defect-based color centers created by radiation, or
• combinations of these, sometimes augmented by surface coatings or residues.

Each state has different stability, different spatial distribution, and different analytical implications. A jeweler's loupe, a refractometer, and a specific gravity measurement can confirm chalcedony family identity and rule out many simulants, but they generally cannot determine whether a red chalcedony is naturally colored, heat-treated, diffused, or irradiated. Those questions require spatially resolved chemistry or spectroscopy, interpreted alongside microscopic texture.

What Remains Uncertain

Several genuine uncertainties persist. Reference datasets for iron speciation in chalcedony are less comprehensive than those for many crystalline gems, and natural variability is wide, so distinguishing subtle diffusion from natural iron enrichment in a specific specimen can be difficult or impossible with current routine methods. The overlap between natural banding and treatment-induced zoning means that no single pattern is a universal diagnostic criterion.

There is also the fundamental problem of provenance of the color itself: two carnelians with identical bulk chemistry can have different defect histories, and two with different chemistry can look alike. Scientific inference here is necessarily conditional, and a responsible treatment opinion is usually expressed with a confidence level rather than as an absolute identification.

The Central Point

Carnelian is often presented as a simple, uniform red-orange gemstone. In reality, its appearance can be the product of bulk iron-oxide pigmentation, near-surface diffusion, irradiation-induced color centers, or some combination. These are physically different states that share one visual outcome, and they can only be separated by combining spatially resolved chemical and spectroscopic evidence with careful microscopic observation. The same stone may behave differently from another—when heated, fractured, or sectioned—because the color is not in the same place, or is not made of the same thing. Recognizing that distinction is the difference between cataloging a gemstone and understanding it.