In this study, we investigated the RF heating effects of iron oxide, especially Fe₃O₄, using electromagnetic and thermal simulation software.

Electromagnetic simulation of specific absorption rate (SAR) and temperature elevation in subjects scanned with MRI have been studied for many years, with previous introduction of a ‘cumulative equivalent minutes at 43°C’ (CEM₄₃) tissue damage threshold to ensure MR safety¹. Despite many reports of skin burns in the region of tattoos, there are few safety studies concerning RF heating caused by tattoos^2-4. Manufacturers of tattoo ink are numerous and all use a range of dye ingredients, making it difficult to assess the electromagnetic properties of each specific ink pigment. Nevertheless, the possible cause of a skin burn during an MR scan has been reported to be related to iron oxide, which is potentially magnetic and an electrical conductor³. Iron oxide exists in different chemical states: Fe₃O₄ is a known ingredient of black ink pigment, along with Fe₂O₃ and FeO for red and yellow, respectively⁵.SEMCAD X (Speag, Switzerland) was used to calculate SAR, CEM₄₃ values, and to solve the Arrhenius model for thermal change. A single loop surface coil was modelled with four equally spaced capacitors and with a diameter of 100 mm and a width of 10mm. The coil was positioned 5 cm away from the Duke human model⁶ with a resolution of (1mm)³ and tuned for 7 T (S₁₁ < -25 dB) (Fig. 1). An anchor-shaped tattoo was modelled 1mm under the skin layer in the region of the cervical spine using iSEG (ZMT, Switzerland) with longitudinal and lateral lengths of 150 mm and 100 mm, respectively. The model in the region of the tattoo shape was adjusted until a single layer of tattoo pigment was located 1 mm under the outmost skin layer in the 3D slices. The maximum local SAR for normal skin was calculated for a standard permitted thermal simulation input power (10gSAR_max=9.8 W/kg at 43 W input power). The dielectric properties of black tattoo ink pigment have not been reported, so the concentration of Fe in black tattoo pigment⁷ was used to estimate its dielectric properties (σ: 1275.2 S/m, ε_r: 74.41 in Fig. 2). The thermal properties of tattoo ink pigment were assigned from previous studies⁸. To estimate the skin burn effect, the Arrhenius model was used and its parameters for skin were assigned from literature values⁹ (A: 2.2x10¹²⁴ 1/s, ΔE=187.1 kcal/mol). RF heating was calculated during 20 min of simulated surface coil transmission with 43 W input power. Regional temperature elevation was predicted in the bottom of the tattoo region, especially at its edges (Fig. 3). A CEM₄₃ simulation was performed and showed that the left side edge of the tattoo was heated the most (CEM₄₃=209.8 for 10 min, CEM₄₃=409.5 for 20 min, respectively, see Fig. 4). It is expected that heating of between 480 and 960 CEM₄₃ would cause an immediate superficial burn¹⁰. The Arrhenius tissue damage model was calculated to estimate the potential severity of the skin burn, and showing an increase to 0.3 Ω and 0.51 Ω for 10 min heating and 20 min heating, respectively, in the case of a sharpened anchor shape (Fig. 5).The RF heating safety of tattoo ink pigment was investigated. A possible source of skin burn could be caused by a specific form of iron oxide, Fe₃O₄, which is known to have 10⁶ times higher electrical conductivity than Fe₂O₃ due to electron exchange between the Fe^II and Fe^III centers^11,12. Fe₃O₄ is also known as CI pigment black 11, which is insoluble in water and may have impurities, so caution should be used in estimating the dielectric properties of tattoo ink pigment without direct measurement. RF modelling predicted a possible skin burn in an area close to the RF coil copper strip, observed as locally generated heating at the edge of the tattoo shape. For the Arrhenius model, 0.53 Ω is known equate to a 1^st degree burn, and 1 Ω is reported to equate as a 2^nd degree burn¹³, so more heating could lead to the subject suffering a severe skin burn. Future studies are needed to evaluate different tattoo pigment shapes and more realistic dielectric properties of the ingredients. In summary, a simulation model of RF heating in tattoo pigment is proposed, which shows that certain tattoo pigments may lead to severe skin burns when performing high field MRI.