Grass and coworkers have developed synthetic DNA “fossils”, mimicking the inherent hermetic diffusion barrier of fossils.^17,19,46) DNA was encapsulated into amorphous silica instead of “natural’ amber (polymerized terpenes). Silica is a ceramic material with high melting point as well as exceptional barrier property. It can be both synthesized and disintegrated at room temperature under appropriate chemical conditions. Considering the characteristics presented above, silica is an ideal material for an encapsulation-release system to achieve the DNA code system.

Figure 2A(a) illustrates the method for encapsulating DNA molecules into amorphous silica spheres to form reversible three-dimensional (3D) DNA@SiO₂ hybrids. The encapsulation comprises the following procedures: 1. SiO₂ particle preparation, 2. ammonium functionalization of SiO₂ particles, and 3. DNA encapsulation. The prepared silica particles are functionalized through a reaction with N-trimethoxylsilylpropyl-N,N,N-trimethylammonium chloride (TMAPS) to carry positive surface charges as NH⁺. During encapsulation, DNA is adsorbed onto positively charged support particles and a 12-15 nm thin silica layer is formed as a top layer via polycondensation of tetraethoxysilane (TEOS) through the sol-gel process. The structural features of the prepared DNA@SiO₂ were revealed by TEM analysis (Fig. 2B(a)).^17,19)

In the case of DNA@SiO₂, DNA molecules are hermetically sealed within the glass spheres and protected from chemical and physical attack to form stable “synthetic fossils”. If shielded from light, the encapsulated DNA@SiO₂ nanohybrids can be stored safely at room temperature. For further analysis of the protective properties of the encapsulation particles, they are treated with radical oxygen species (ROS) for 10 min, for which the protection efficiency should be > 90 % (Fig. 3A). From the sequencing chromatograms, the cipher DNA released from DNA@SiO₂ nanohybrid even after temperature-treated at 120°C for 15 min showed the same results as that of the original DNA strands. The thermal stability can be demonstrated through the analysis presented above.

For practical application as a code system for information storage, DNA molecules are required to be released from theencapsulates and retrieved without harm for the decoding process. The DNA code can be recovered from the glass spheres without damage using fluoride-containing buffered oxide etch solutions. After dissolution of the spheres and purification, the DNA codes can be analyzed by standard techniques including gel electrophoresis, sequencing, and the quantitative polymerase chain reaction (qPCR).

LDHs, which are also well known as anionic clays, are represented as a two-dimensional (2D) layered crystal structure with a general chemical formula of [M²⁺_1-xM³⁺_x(OH)₂]-(A^m−)_x/m·nH₂O, where Mⁿ⁺ is a metal cation and A^m− represents an interlayer anion. LDHs are composed of cationic metal hydroxide host nanosheets, where some of the M²⁺ sites are partially substituted with isomorphic M³⁺ ions, in order for the host layer to acquire a positive charge. In order to satisfy the charge neutrality condition, exchangeable anionic species should be stabilized in the interlayer spaces of LDH hosts. LDHs are ideal materials for hybrids owing to their advantages such as variable chemical compositions, anion-exchange capacity, host-guest interactions, crystallization dissolution characters, and biocompatibility. ^50-55)

Negatively charged DNA molecules are immobilized inside the positively charged LDH matrix to achieve a stable DNA-LDH nanohybrid system. The intercalation chemistry for DNA-LDH nanohybrids was proposed for the first time in 1999 as “bioinorganic nanohybrids”.⁶⁾ In 2004, pioneering studies on DNA-LDH hybrids for information storage were conducted by Choy et al.¹⁶⁾ The typical 2D layered DNA/LDH hybrids were prepared by an anion-exchange route and demonstrated their applicability as a genetic molecular code system.

For bulky and large molecules, the above encapsulation method often results in low availability because anion adsorption to the LDH surface is restricted by physical limits on lattice expansion. As shown in Fig. 2A(b), a new approach termed “exfoliation-reassembly” provided a synthetic pathway to stabilize the DNA code inside a spherical core, reassembled with LDH single nanosheets, to form a 3D DNA@LDH core-shell.⁵⁾ A nitrate-intercalated LDH was prepared first, then delaminated into single [Mg₂Al(OH)₂]⁺ nanosheets in a suitable solvent. The colloid of exfoliated LDH single sheets was then blended with the aqueous solution of DNA, where LDH nanosheets were utilized as a building block and DNA as a charge-balancing anion guest species. The electrostatic interactions resulted in self-reassembly to achieve a chemically well-defined and uniformly spherical DNA@LDH nanocapsule. The features of the prepared DNA@LDH core-shell structured hybrids were revealed using TEM analysis (Fig. 2B(b)).

With respect to DNA@LDH, in the first encoding step, a sequence-manipulated double-stranded DNA (dsDNA) of a 35-base-pair duplex conjugated with green-color Cy3 or red-color Cy5 fluorophore was designed as a molecular tracer. In the second encrypting step, the designed DNA code was encapsulated in the core space of the LDH nanoshell with an overall thickness of ≈ 10 nm to produce a DNA@LDH bio code with a particle size of ≈ 100 nm. It is suitable for applying the bio code into various products. The LDH surface can be modified easily, and it is possible to apply the nano code in hydrophobic fluidic systems. In the third decrypting step, a small amount of DNA@LDH nanoparticles was collected, and DNA strands inside the LDH layers were safely released upon the pH-dependent dissolution of LDH. These can then be recovered through appropriate methods (e.g., magnetic separation) for analysis. In the final decoding step, DNA or the PCR product after amplification could be identified by DNA-chip assays or gel electrophoresis.

Stabilization of the DNA codes within the nanosized LDH matrix is one of the most important abilities for identification systems. The results of a thermal stability test, which detects melting points and chemical stability against copper-catalyzed DNA damage, showed the superior protection ability of DNA@LDH nanohybrids. In addition, stability against enzymatic damage is the most significant benefit of DNA@LDH hybridization. As shown in Fig. 3B, even after treatment with DNase I, a kind of ribonuclease, both the DNA released from LDH and the free DNA exhibited the same electrophoretic mobility. This provided convincing evidence that the structural features of DNA were fully protected from DNase I treatment.

Combining the strategies presented above, both DNA@SiO₂ and DNA@LDH bio codes ensure a high level of identification and provide advanced information storage with confidentiality and inherently high security.

Global counterfeiting and adulteration extend to all types of products, especially high-profit and high-volume merchandise. Fuels, high-value cosmetic products, medicines, foodstuffs, and even food are routinely counterfeited for large illegitimate profits. This causes serious societal instabilities, e.g., large monetary losses for governments or a great threat to consumer health. Determining the authenticity of products is fundamental for both consumer health and trader interests.

Grass et al. have proposed methods for oil authentication and tracing a food product along its production chain. In 2014, they applied silica particles with encapsulated DNA (SPED) as both, invisible oil tags and unique food tracers.^20-21) Silica particles have long been used as food additives (E551), and there is immense potential for applying the SPED tracer to various food types.

A successful application of the SPED tagging/tracing technology in three model oils, a fuel oil (gasoline), a food-grade oil (extra virgin olive oil), and a therapeutic oil (bergamot oil) has already been demonstrated. A thermostable, hydrophobic, and magnetic DNA@SiO₂ (Fe₂O₃/TMAPS/DNA/SiO₂-C6) particle has been developed for use as an invisible tag for oil products. The barcode function of the inert taggants was used in order to prove the authenticity of the oil-based items.

For encoding, dsDNA of ≈ 100 base pairs was purchased by Microsynth AG. In the encrypting step, Fe₂O₃/TMAPS/DNA/SiO₂-C6 particles were synthetized, with a core-shell structure and consisting of magnetically recoverable nanoengineered shells embedding artificial DNA sequences (Fig. 4A(a)). Fe₂O₃ nanoparticles were synthetized by an inexpensive synthesis method based on co-precipitation and were then further functionalized with ammonium groups using TMAPS. DNA molecules were adsorbed on the prepared magnetic particles, and the Fe₂O₃/TMAPS/DNA nanocluster was coated with a dense SiO₂ layer using TEOS as the Si source via sol-gel chemistry. The silica surface was further functionalized by hydrolysis of N-hexyltrimethoxysilane to achieve dispersibility in hydrophobic fluids. The hydrodynamic particle size of the obtained nanoengineered encapsulate distribution in toluene was ≈ 135 nm, and the tags did not affect the optical properties of the dispersant at concentrations below 103 μg/L. The thermal stability of DNA encapsulated within the particles was demonstrated in comparison to unprotected DNA. The taggants could still be detected after being treated at 160°C for 30 min, while naked DNA was no longer detectable after being heated in water to 120°C. The additional test for long-term storage capability also showed the good stability of the prepared materials in decalin at room temperature for durations of up to 2 years.

Encrypting is the key process for tagging (Fig. 4A(b)). Therefore, it is important to collect Fe₂O₃/TMAPS/DNA/SiO₂-C6 particles from oils efficiently and recover the DNA from encapsulates without damage. Using magnetic separation, even low concentrations of particles (down to 1 μg/L) and low volumes (1 mL) could be collected from oil systems for further treatment. Both the Fe₂O₃ and SiO₂ matrix can be dissolved in buffered oxide etch (BOE) solution. The recovered DNA molecules were analyzed by qPCR or Sanger sequencing for decoding (Fig. 4A(c)). The novel magnetic DNA@SiO₂ particles can be utilized as low-cost (2 × 10⁻⁴ USD/L) tracers, and the method is universal for a variety of oils and oil-derived products.

The inert taggants, which have stable core-shell structures, were composed of an iron oxide core and a protective silica layer. Each of the ceramic materials plays a significant role in the encapsulates, in which the Fe₂O₃ core confers the magnetic properties for separation, allows for sample concentration, and facilitates handling, while the surrounding SiO₂ matrix can act as hermetical sealing layer and is responsible for heat stability, surface functionality, as well as recoverability for the DNA tags.

The SPED were also developed as food tracers to a milk processing chain (Fig. 4B). The food labelling technology along its production chain is essential to ensure product authenticity and food safety. Both synthetic (MB1) and naturally occurring (TOM1) DNA sequences were utilized as DNA codes. The SPED were prepared by the approach introduced in Section 2-1 and the stability of the DNA code protected within SPED was demonstrated. In this study, the developed particles were added to milk, as well as milk-derived yoghurt and cheese. Following extraction of the tags from the food matrix, upon mixing with BOE solutions, the silica layer can be dissolved and the released unique DNA code could sufficiently be quantified via quantitative qPCR at tracer concentrations as low as 0.1 ppb (μg per kg milk). The results have demonstrated the potential of SPED as a sufficiently stable and precise food tracer.

The DNA@LDH core-shell structured hybrid, which is one of the DNA@ceramic steganographies introduced in Section 2.1, can be used as a highly specific molecular coding system. DNA@LDH nanoshells were developed by Choy et al. in 2010.⁵⁾ The LDH ceramic shell plays an important role in the system to provide efficient protection for DNA code, robust support during data processing in various kinds of media, as well as on-demand de-capsulation. DNA@LDH bio codes have received more attention due to the advantages of LDH in identification systems including highly precise nano-forensics. A completely new convergence science, called “nano-forensics”, has developed with advanced nanotechnology merged with forensic science. In nano-forensics, nanoparticulate traces are utilized to investigate issues such as the adulteration of chemical products, document copyright counterfeiting, drug trafficking, terrorists, and so on.

As a proof-of-concept, the authors validated that the DNA@LDH nanohybrid was sufficiently useful in a DNA-based document authentication system (Fig. 5). A security element [S]-shaped signature was printed on a glass slide with a nanohybrid-suspended black ink prepared by integrating the DNA@LDH nanohybrid into conventional ink. The DNA@LDH nano-forensics molecules were created through the strategy described in section 2.1. Although both [S] characters printed with the nanohybrid-formulated ink and pure ink look the same to the naked eye, one can differentiate the secured forensic [S] character from the nonforensic [S] character by fluorescence analysis. The nanohybrid-suspended [S] marking shows yellow-merged green and red fluorophores labeled in the designed DNA code. For further accurate identification of base sequences, the DNA molecules were decrypted from the [S] signature by resolving the LDH shell in a slightly acidic solution (pH = 4), along with the addition of EDTA, a chelating agent for trapping Mg²⁺ and Al³⁺ ions. The released forensic DNA code was collected and then detected by the microarray. The DNA-chip (12 × 12 array, 144 spots) could be pre-patterned as the [S] shape in green/red, in which each spot was designed with complementary sequences to those in DNA@LDH. The released dsDNA from the hybrid was denatured to single strands after heating at 95°C for 2 min, which were detected by DNA-chip hybridization in one second. Hence, the security DNA information was successfully identified as an [S] graphic of representative spots.

More recently, Choy et al. combined the above mentioned methods in the DNA@LDH nanohybrid bio code with an electrical mobile smartphone device to suggest a new concept called “Avatar DNA”, as illustrated in Fig. 1A.³⁾

Avatar DNA is an optical machine-readable icon on a DNA biosensor for automatic identification and data capture (ADIC) in the form of Quick Response (QR) and Color-Zip codes. Therefore, specific information encoded in a DNA base-sequence can be identified through a photographic scan from a smartphone, and can connect offline DNA image codes to online remote databases, including secret messages, indexes, profiles, and URLs.

As a proof-of-concept experiment, the Avatar DNA platform was applied to strengthen the supply-chain security of an injectable cancer medicine, Avastin. The comprehensive item-level identification system was elaborated by four steps: encoding, encrypting, decrypting, and decoding. In the encoding step, the secured AVASTIN DNA codes were encoded using binary coding via ASCII characters, the base-3 Huffman-compression coding method and a practical DNA encoding method. The encoded information was expressed as a 2D QR or 3D ColorZip code through a logical code algorithm (Fig. 6A). In the encrypting step, a DNA@LDH hybrid bio code was prepared as mentioned in Section 2.1, and artificial DNA molecules with red, green, and blue color fluorophores were encapsulated in LDH ceramic shells to produce a series of three differently colored nanohybrid-suspended inks. The DNA-QR code and the DNA-ColorZip code pattern were printed with the prepared nanohybrid-suspended inks in the form of a packaging label (Fig. 6B). In the decrypting step, fluorescent Avastin DNA-AIDC images were captured and read by scanning with a smartphone camera to display the hidden DNA codes on the screen. Thus, the users can check timely intermediate results by directly connecting offline DNA-AIDC tags with an online reference database. In the decoding step, the AVASTIN DNA molecules were collected by the routes described in Section 2.3, and DNA identification proceeded through hybridization on each DNA-chip. The resulting DNA-QR icon and DNA-ColorZip icon on the microarray were automatically identified by a smartphone with the required application, which can provide further information through wireless networking, namely the phone-based tele-info communication.

Therefore, the NBIC convergence strategy, in which each building block stands for ceramic nanosheets, nucleotides, digits, and pixels, could sufficiently verify the item-level identification that strengthens supply-chain security against counterfeits.

Tuning of the lifetime is a method that utilizes the temporal domain of the luminescence. Fluorescence lifetime encoding with rare earth metals, which have a long lifetime ranging from micro to milliseconds, is preferred over organic dyes and QDs, which have shorter lifetimes (organic dyes < 10 ns and QDs: 10-100 ns).^26,60)

Jin and coworkers have shown that tunable luminescent lifetimes, referred as “τ-dots” in the microsecond region, could be utilized to code individual UCNs.²⁶⁾ In this study, energy transfer from the sensitizer Yb³⁺ to blue-emitting Tm³⁺ ion at changing sensitizer-emitter distances allowed lifetime tunability. As shown in Fig. 7B. by modulating the Tm³⁺ ion concentration between 4, 2, 1, 0.5, and 0.2 mol% with a fixed 20 mol% of Yb³⁺ ions in 40 nm NaYF₄:Yb,Tm nanocrystals, the life time could be tuned from 48.6 to 668 μs (Fig. 7B). NaYF₄:Yb,Tm nanocrystals were synthesized by the thermal decomposition to an average particle size of approximately 40 nm with a narrow size distribution (Fig. 7A). Eight distinct lifetime codes were prepared by a new temporal-domain approach, and more distinguishable codes could be produced by fine-tuning the co-doping ratio of Yb/Tm.

As a proof-of-concept, it was demonstrated that the purpose-engineered τ-dots carrying lifetime codes can be used as barcodes in document security for anti-counterfeiting. Fig. 7C presents three overlaid images of the Macquarie University logo, Sydney Opera House, and Sydney Harbour Bridge, printed with τ-dots-suspended commercial ink with Yb:Tm doping concentration ratios of 20:4, 20:1, and 20:0.5 (mol%), respectively. The overlapping patterns (left) printed with NaYF₄:Yb, Tm nanoparticles with three distinct lifetimes could be decoded (right) by time-resolved scanning cytometry with a 980 nm laser. Three individual images, the Macquarie University logo in red having a lifetime of 52 μs, the Sydney Opera House in green having a lifetime of 159 μs, and the Sydney Harbour Bridge in blue having a lifetime of 455 μs, can be clearly observed. This result suggests a new type of document encryption with τ-dots as security inks. The prepared nanocrystals have also been used to create a new matrix of τ-dots-encoded microspheres as multiplexing suspension arrays.

When upconversion nanoparticles were used as an optical code for document security and anti-counterfeiting, the emission color was the basic security element, and was tunable from visible to NIR light. Increasing the number of emission colors and manipulation of morphology are both straightforward and efficient strategies for meeting the criteria for multilevel anti-counterfeiting applications.

Doyle and coworkers reported a robust encoding strategy that combined distinct UCNs with spatial patterning and the generated unique polymer microparticle barcodes exhibited high-density encoding (> 10⁶ particles) and an ultralow decoding false-alarm rate (< 10⁻⁹).¹⁴⁾

An initial set of UCNs 1-9, which were bright and unique upconversion nanoparticles, were synthesized through the hydro thermal reaction by adjusting the relative dopant concentrations of the ions Yb³⁺, Er³⁺, and Tm³⁺ in the reaction premix, which could be distinguished by the naked eye after a 980 nm NIR excitation. Fig. 8A illustrates integration of UCNs into polymer microparticles using stop-flow lithography (SFL). Two different kinds of particle monomer chemistries were explored, namely poly(urethane) acrylate (PUA) for temperature- and chemical-resistant microparticles and poly(ethylene glycol) diacrylate (PEGDA) for biocompatibility. Integration with hydrophobic PUA or hydrophilic PEGDA particles requires rational surface chemistry. The monomer solution (PUA or PEGDA) with surface-modified UCNs was photopolymerized through illumination with mask-patterned ultraviolet light (365 nm). By embedding distinct UCNs within polymer microparticles to form a stripe-like emission feature consisting of up to 6 stripes, encoding capacities in excess of 1 million can be achieved. This high-throughput particle synthesis provided stripe-like patterned multicolor barcodes with a single-particle encoding capacity of C^S for asymmetric particles and C^S/2 for symmetric particles, where C is the number of UCN “colors” and S is the number of microparticle “stripes”. This intuitive coding motif scale can increase rapidly with incremental changes to either distinguishable spectral signatures—UCN “colors” by adjusting Yb³⁺/Er³⁺/Tm³⁺ ratios with no impact on decoding, or spatial features — microparticle “stripes” by the additional stripe.

Figure 8B and Fig. 1B (top) show the extraordinary flexibility and practical utility of this universal architecture when utilized as covert labelling barcodes over a range of complex products including pharmaceutical packaging, credit cards, curved ceramic objects, reproduced artwork, high-temperature-cast polystyrene, and currency. The durable anti-counterfeiting labels can withstand extreme conditions of plastics engineering processes such as high-temperature casting and lamination. The covert labelling with a virtually unlimited encoding capacity of (C^S)^N for asymmetric and (C^S/2)^N for symmetric particles (N: the number of particles deposited) can efficiently identify an object.

Notably, a straightforward portable charge-coupled device (CCD)-based decoding was presented here. The portable decoder consisted of a mobile phone adapter (Apple iPhone 4S) and a × 20 objective lens. Images of prepared stripes-like barcodes were captured using the portable CCD decoder and loaded into a software (Mathworks, Natick MA) to be defined using the detection algorithm. Implementation of quantitative decoding with the designed portable apparatus was straightforward as it utilized the same central components (CCD detector and magnification) as those of the microscope-based instrumentation. It was demonstrated that the covert multiparticle barcoding could be decoded readily even in complex surface conditions such as a blister pack or curved surface.

More recently, versatile nanoplatforms with a core-shell structure combining emission color and lifetime modalities have proven especially useful for high-density information storage and multilevel anti-counterfeiting. The precisely controlled core-shell structure can induce excitation under various irradiation wavelengths.^27,61)

In 2017, Liu et al. developed binary temporal codes for lifetime-based data encoding by integrating the long-lived upconversion emission of transition metal ions Mn²⁺ and relatively short-lived lanthanide emission decay.²⁷⁾ With well-controlled core-shell structures, the nanoparticles could be excited under both 980 nm and 808 nm irradiations. The prepared Mn²⁺-doped nanoparticles allow a high throughput rate of authentication, and the emission decay was distinguishable even by the naked eye without using any time-gated decoding instrumentation.

The multilayer nanoparticles were designed to have a long-lived upconversion core and short-lived upconversion shells. Hexagonal-phase NaGdF₄:Mn (30 mol%) and NaGdF₄: Yb/Tm (49/1 mol%) are generated as the core and first shell layer, respectively, to enable the long-lived Mn²⁺ upconversion emission. Considering that the lifetime of a Mn²⁺ emission (39 ms) is approximately 65 times longer than that of a Tm³⁺ emission, the emission can be detected by the naked eye. In addition, the core-shell structure of multilayer nanoparticles can also contribute to retaining the long lifetime of Mn²⁺ emission by avoiding the back-energy transfer from Mn²⁺ to Yb³⁺ due to separation. Succeeding layers NaYF₄ or NaYF₄:A (A: Nd, Eu, Eu/Tb or Tb) can be conveniently passivated in support of tuning the short-lived lanthanide emission color, while preserving the long-lived emission of Mn²⁺.

The multilayer nanoparticles were synthesized as follows (Fig. 9A). First, hexagonal-phase NaGdF₄:Mn (30 mol%) core nanoparticles were prepared by a hydrothermal method. Next, epitaxial growth of NaYF₄:Yb/Tm and NaYF₄ layers onto the cores was enabled by a coprecipitation reaction at 290°C for 1.5 h, followed by thermal decomposition at 290°C for 0.5 h. The TEM image in Fig. 9A shows the upconversion nanoparticles with a particle size of 17 nm. These were further doped with Eu³⁺ or Er³⁺ in the presence of a NaYF₄ inert shell layer to enrich the variety of the short-lived upconversion emission colors. The epitaxially grown shell of NaYF₄:Nd (20 mol%) on NaGdF₄:Mn (30 mol%)@ NaGdF4: Yb/Tm (49/1 mol%) nanoparticles could generate Mn²⁺ upconversion emission at 808 nm because of the energy transfer from Nd³⁺ to Yb³⁺.

As a proof-of-concept, the authors demonstrated that the Mn²⁺-doped multilayer nanoparticles were suitable for multilevel anti-counterfeiting applications on reproduced artwork as binary temporal codes. 2D-converted patterns were stamped by Mn²⁺-activated core-shell nanoparticles of different composition and colloidal NaYF₄:Yb/Er (20/2 mol%) nanoparticles as a control (Fig. 9B). Steady irradiation at 980 nm (6 W cm⁻²) led to multicolor patterns, while dynamic scanning of the identical patterns with a focused laser beam (64 W cm⁻²) yielded a different result. Unlike only a main green spot of emission emerging from the pattern stamped by purely lanthanide-doped nanoparticles, the as-prepared multilayer nanoparticles showed a bright main spot of emission with an additional tailed emission in green. The color of the main spot was varied with different doping compositions. Additionally, patterns prepared with Nd³⁺-sensitized nanoparticles were found to have similar emission features under excitation at either 980 nm or 808 nm.