Optical memory enters 5D realm
Glass nanostructures etched using high-intensity femtosecond laser pulses promise to keep vast quantities of data safe for billions of years, describe Peter Kazansky, Ausra Cerkauskaite and Rokas Drevinskas
Our ability to store and access data has grown rapidly during the 21st century. The Internet is increasingly bringing all forms of information technology to everyone’s fingertips, making life faster, more informative and more connected than ever. However, with individuals and organizations generating ever-larger datasets, we desperately need more efficient forms of data storage that have high capacity, low energy consumption and a long lifetime.
Despite immense technological progress over the past few decades, it is still difficult to store lots of information securely over even relatively short timescales of 100 years. Data stored in traditional magnetic systems, such as tapes and hard disks, have to be transferred every couple of years to prevent them being lost as the drives start to wear out. As for conventional optical discs such as CDs and DVDs, they might last a few decades before the reflective layer starts to corrode or the disc material itself breaks down. The third main technique for data storage – solid-state devices based on semiconductors, which underlie flash and solid-state drives – provides an even shorter lifespan of barely 10 years because transistors become unreliable after a number of program/erase cycles.
The explosion in digital information is a record of 21st-century civilization, but we risk losing this data as current storage technologies fall short in capacity and lifetime.
Recently, researchers have made promising progress towards a high-capacity optical memory that lasts not decades but perhaps billions of years. Based on ultrafast laser-induced nano-gratings fabricated in fused quartz, our team at the University of Southampton in the UK has been able to demonstrate a novel “5D” optical memory that promises essentially limitless data storage. The technology, which emanated from work done under the framework of the EU’s Femtoprint project, was first experimentally demonstrated in 2013 and has since been used to record major documents including the King James Bible, Newton’s Opticks and the Universal Declaration of Human Rights.
CDs only have two dimensions in which to store information: tiny pits on the CD surface that either reflect or do not reflect laser light to convey the 1s and 0s of binary data in a single layer of plastic. In DVDs, data are stored by burning pits on multiple layers, adding a third storage dimension. In contrast, 3D optical-storage techniques potentially allow us to write thousands of “layers” in a single monolithic disc without adding a single physical layer.
3D optical storage was first demonstrated 20 years ago by physicists at Harvard University in the US using femtosecond laser pulses to deliver a precise energy to a glass substrate in a tightly confined volume (Opt. Lett. 21 2023). Normally in a single memory cell or voxel only one bit of data can be stored, but nonlinear optical effects allow the cell dimensions to be shrunk by a factor 10, therefore increasing capacity.
The latest developments in 3D optical memory have enabled a feature size below 100 nm with an equivalent capacity of approximately 10 TB per disc by using a dual-beam technique named super-resolution photoinduction-inhibition nanolithography (Opt. Lett. 36 2510). This technology could let us break the diffraction barrier, and in 2013 was used by a team in Australia to achieve feature sizes down to 9 nm, offering a capacity of up to 30 TB per disc (Nat. Commun. 4 2061). For comparison, a standard DVD has a capacity of just 4.7 GB.
To increase the data capacity of optical storage, there is the potential of storing more than one bit in a single voxel using multiplex technology. During the past few years, several parameters such as polarization, wavelength, space and fluorescence have all been trialled to act as the additional dimensions for optical data storage. Various materials have also been implemented for multidimensional data storage, such as silver clusters embedded in glass and gold or silver nanoparticles. The recently developed 5D optical-storage technique, in contrast, uses birefringence as an extra degree of freedom – the property of a medium whereby its refractive index varies depending on the polarization and direction of incident light. Birefringence generated by the orientation and size of optical nano-gratings offers two extra dimensions, thus providing much higher storage capacities.
The discovery of 5D optical data storage was serendipitous. Our group, led by one of the present authors (PK), had been collaborating with a group at Kyoto University in Japan for the previous decade exploring what happens when intense light fields produced by powerful short-pulsed lasers strike certain materials. Although the original aim was to study the light–matter interaction and other fundamental optical phenomena, in 1999 the team realized that a self-assembled nanostructure in silica glass was being formed by a tightly focused linearly polarized beam.
First observed microscopically in 2003 using fixed, focused femtosecond laser pulses, these self-assembled nanostructures – measuring roughly 20 nm across – are among the smallest embedded structures ever produced by light (Phys. Rev. Lett. 91 247405). Despite several hypotheses to explain the physics of the peculiar self-organization process, the formation of the nanostructures remains under debate. However, once the extraordinary stability and optical properties of anisotropic femtosecond laser direct-written structures were identified in 2006, it turned out that this unusual phenomenon would be useful for multiplexed data storage.
The first demonstration of storing visual information in five dimensions was in 2010 (Adv. Mater. 22 4039), marking the first step towards practical nanograting implementations. The following year the technique found applications in commercial polarization converters, and by 2012 it had started to hit the headlines. The Telegraph newspaper dubbed it “Superman” memory, and Hitachi announced 3D glass data storage that will last for millions of years. Following the high public and scientific interest, in 2013 our group demonstrated the first ever digital data recording. Since then, a number of different types and volumes of data have been stored.
Although there have been several promising alternative attempts to develop long-term and high-capacity data storage that lasts for millions or billions of years, many of these – such as DNA storage, silicon-nitride/tungsten-based media and microscopically etched/electroformed nickel plates – are expensive and too slow to be practical. 5D optical memory, in contrast, is far superior, especially when applied to fused silica, which has a high chemical and thermal stability. The lifetime of 5D memory is 1020 years at room temperature, indicating unprecedented stability among all techniques (Phys. Rev. Lett. 112 033901). In addition to the benefits of multiplexing, 5D optical data based on nanogratings can also be erased and rewritten – two key features when considering data storage.
The current data-writing system is not much different from that found in CD or DVD drives. Ultrashort laser pulses with a wavelength of 1030 nm are focused inside a spinning glass disc and the position, power and polarization of each pulse are simultaneously modulated depending on the encoded information – leaving a trace of pits with different optical characteristics. Reading the data is more complicated because it requires a microscope-based birefringence measurement system, but we are now working on how to solve this problem.
Our 5D technology has so far only been demonstrated in a laboratory setting. The main bottleneck is its slow writing speed (roughly 100 bit/s), but we are striving to increase the rate and to develop a microscope-free readout drive. This requires a high-speed polarization controller and industrial-grade disc drives, and we are currently looking for potential partners and investors. A number of large organizations that need to archive lots of data, including Microsoft, Sony and Warner Bros, have already contacted the group. Libraries, hospitals and governments could benefit from our approach, too. With support from investors, commercial 5D data storage could be achieved within several years.
Another vital step is to drastically increase the capacity of the storage. By recording data with tighter focusing optics and shorter wavelength light, it is possible to achieve a spatial (3D) densification similar to that in Blu-ray discs, involving a pit size of less than 200 nm. Combined with the fourth and fifth dimensions provided by birefringence, which allow a single pit to store eight bits (one byte) of information as opposed to one, it would be possible to achieve an unprecedented capacity of hundreds of terabytes in a single 12 cm-diameter disc. Perfecting 5D storage technology would therefore be a major step towards preserving the digital age for future generations.