Title: The micro-design of hooked biological attachment mechanisms and soft robotics – a Biomimetic approach.
Hooked attachment mechanisms are a subset of all Biological Attachment Mechanisms and a useful starting position for experiments on the imaging of all biological attachment mechanisms such that they can be adopted in the engineering domain. A hook has an overhang which makes the imaging and transfer to .stl format a challenge, a test that once passed, allows for the further imaging of attachment mechanisms of all shapes and of differing materials. Confocal microscopy seems to have solved the issue so that it is now possible to move from the attachment mechanism directly to the finished model without user interference . Here, the work to-date is summarised, imaging cellulose and chitin hooks so that the process can move forward to other attachment devices of interest such as the mating parts of sexual organs in insects or other biological sub-structures that are not hooked. Progress is reported to have been made into the development of chitin nano-tubules so clearly there is hope that this work will yield a standard for mechanical attachment mechanisms of soft tissues or materials that can interact safely with human flesh with medical applications.
Keywords: hooks, probability, scaling effects, biomaterials.
This is a review article of the three papers published in the Springer-Open journal, “The Journal of Robotics and Biomimetics” in a special issue on nano-/micro-robotics under the following titles:
1. A biomimetic study of natural attachment mechanisms— Arctium minus part 1 
2. A biomimetic study of natural attachment mechanisms: imaging cellulose and chitin part 2 
3. Micro-design using frictional, hooked, attachment mechanisms: a biomimetic study of natural attachment mechanisms—Part 3 
The title of part 3 above displays the underlying motive behind the exploration of the detail of papers 1 and 2. It accepts the viability of using cladistic methods to arrive at a scenario where a structure that has survived the “evolutionary sieve” is selected, to quote Nicklaus et al , over the use of Linnaeus or other classification methods which can be seen as insignificantly better when it comes to evolutionary manifestations of properties and/or structures.  goes some way to describing this technology transfer.
The first view was that it was unsuitable to study with available technology. The decision was made to proceed with the use of a confocal microscope instead of light microscopy. Subsequently it has become possible only through the work of Hirt et al , by their work on a layered manufacturing device that can accept .tiff files as input and produce form. Now a hook can be manufactured at a 1:1 scale to the specimen that is to be reverse engineered and that means that designers are on the brink of being able to make things that are of use, in the micro-realm (of the order of 10-100 microns in size). It all began with the discovery that it was possible to image one of the hooked probabilistic fasteners under laser light, namely the cellulose hook of burdock (Arctium minus). Therefore the work continued with the chitinous growths of the bee and the grasshopper (Apis mellifera and Omocestus viridulus) tarsii . This encounter with luck was able to make true the theory that the use of the microscope could be for the imaging of a specimen and then the transfer of data directly to a layered manufacture device that was suitable, namely the work of Hirt et al. The point of this imaging was to use it to describe the group of probabilistic fasteners as a number, namely one for the hook, two for the attachment mechanism of the grasshopper O. viridulus with two hooks, and three for the double set of hooks, namely A. mellifera with a separating arolium, irrespective of component material.
The chance of being on top of a specimen structure available without travelling was immense, as these were all available at the University of Bath which is set in the countryside of Western England. Particularly the burdock which is used (apparently) as the basis of Velcro but it is concluded this is without fundament and it seemed better to use it than to use the others (see below), as it will be shown, for the production of a new hook, a multi-use flat structure of multiple hooks that could be used without being entirely known, as per its value and knowledge. i.e. if it is to be the one to be imitated then it needs to be studied more now so that it can be manufactured.
Figure 1: An Arctium minus (commonly known as burdock) fruit showing milli-metric scale. 
In Part 1 of the investigation , the cellulose hooks of burdock revealed a scaling effect  under loading. This is because the hook un-rolls as it is loaded until the radius of curvature is increased in size at fracture, in a similar manner in which a length of iron chain cannot be horizontally loaded until it is pulled straight without failing. The material is simply not as stiff as it would appear in the sketch of the structure for analytical purposes with its Newtonian assumptions and its properties vary under conditions, such as its state of dessication.
The reasons for this have been considered but not concluded as of yet, requiring further inspection of the material properties. All the natural cellulose hooks studied in the literature, Agrimonia eupatoria, Circaea lutetiana, Galium aparine, and Geum urbanum as well as Arctium minus, have been described in terms of their originating structures .
Table 1: Grouping the cellulose, probabilistic, frictional and long-shafted hooks according to originating structure.  and .
The cellular complexity obviously plays a part and from  the micro-fibril strengthening of the structure must play a part too, but this does not satisfy the Newtonian equations of static analysis used for hooks of a larger size. This is an exciting find since it suggests that there may be differing laws governing the behaviour of structures at this level other than standard analysis, rather in the way that the behaviour of fluids differ under different flow conditions  governed by the energy equation. Therefore the sense is that it is best to mimic the morphology exactly in order to yield optimal performance and maximum attachment strength when fastened, through fiction and mechanical attachment, bearing in mind that a hook must be paired with a substrate.
The aim therefore of  through  was to develop a methodology whereby a Universal micro-robotic frictional probablistic attachment mechanism can be derived such that its performance can be modelled graphically, using Biomimetic principles and such that the methodology can be applied to other, more complex attachment mechanisms in the future. It is called a Universal Foot after the fact that a human foot is a frictional probabilistic attachment mechanism and because its performance is to be modelled graphically for design, performance, material, quality and other parameters, its universal qualities.
Arctium minus is Class 0. Using copper, cheap and therefore available to mass production, it regarded to be the best fit for the solution of making a reproduce-able hook that will sustain in making it to the end of the product lifecycle. See  again for the details of the imaging and deposition process. A sample hook was placed under a single phase confocal microscope and recorded (see Figure 2). It was digitised and loaded into Solidworks (c) and analysed (see Figure 3).
1. 2. 3.
4. 5. 6.
7. 8. 9.
10. 11. 12.
13. 14. 15.
16. 17. 18.
Figure 2: 1 – 20 The individual z-axis scan .tif files that make up the stereogram of the burdock hook (the scale bar defines 200 microns, Dr I Jones October 2002). 
With respect to a Universal Foot it is impossible to measure its probability of fastening since there is a possibility that it may not hold the correct angle on the surface/substrate. That will be overcome with a hinge that will allow the foot to align with the ground according to its angle and not the angle of application. It therefore can be used to develop further since it has application to the frontier of technology and the use is yet to be completely foreseen, such as soft robotics, micro-robotics, biosensors, computer hardware, orthodontics and optical sensors through the use of copper which is a very known substance with qualities that have been researched and ascertained through its use as a strain gauge and other common applications, and its coating of biologically inert stainless steel.
It will be seen that there are a number of solutions to the problem of a Universal Foot and that means a testrig will have to be devised such that it can measure the forces with which a hook attaches to a substrate and that is the way through to the end of the series such that each member of the group of probabilistic fasteners can be measured, of different biological materials as imaged in . In the meantime it is possible to make deductions such that a design can be arrived at that resembles a caterpillar yet makes use of the hook of the burdock and the range of movement that requires needful thinking so that it can be measured. Once this is done we have a product which can be commercialised. Part 2  contains the results of the experimentation to image cellulose and chitin and this will prove useful in the future when we consider a wide range of hooking and other mechanisms/devices since it will be in the interest of those continuing the study to know the difference between the two and whether they can use the data to make hooks that are biological such as those to attach to the stomach wall or the vessels of the heart since they bear cilia which makes them difficult to render in a stainless steel as with a stent. But when it is available it may be possible to make them from a biological material which does not dissolve such as the MIT device which, when swallowed, removes a watch battery from the stomach wall to avoid a ulcer forming there or to patch a wound, steered by magnetic fields and which is still in the experimental phase. It is made from pig’s sinew which is insoluble but which does not lend itself to electro-deposition of course so an alternative will need to be found. The electrodeposition of stainless steel has been investigated by Hasegawa et al  and it shows that an improvement has been made to the processing of an otherwise inert steel that does not corrode or “anodize” and it can be electro-deposited on copper. This will make the stainless steel coated copper relatively biologically inert.
Figure 3: The maximum deformation under loading. A point load at the tip, constrained at the base along the flange. There is nothing unexpected about the mode of deflection which reflects static Newtonian loading. This image is constructed using 2-D digitising due to the Nature of the available technology. 
Within the constraints of Nachtigal’s classifications , three hooked classes have been imaged on a confocal microscope  and all that remains is to pass the data to the mechanism of  to produce prototypes for testing. In a manner of regard, essentially multiple Class 0 hooks have been assembled in an array as a collective or field (see Figure 4). They are shaped as per the cellulose form of the burdock hook which is simple and shows no stress modifications, with a tapered tip. Manufactured from copper, their attributes have yet to be discovered but it is hoped that it will yield an attachment device that will succeed in vertical assent via quadrupedal locomotion. It will be designed to be multi-use, temporary and permanent, probabilistic and frictional. Its physical properties will of course differ not least for copper’s well-known capacitance to pass electric current and its magnetic properties.
Figure 4: A zipper configuration in isometric view. This illustrates the possibilities of a composite formation of long-shafted hooks acting a coordinated fashion. The point being illustrated here is that although we are seeking a Universal “foot”, it is as likely to look like a foot as a drone looks like a hummingbird. 
For many years scientists have been studying the work done and methods of doing so in the animal world. The work being energy transfer and the methods, from walking to holding a stone as a hammer. It now has become possible to study the intimate details of the assembly of life and it is also becoming a useful aptitude to be able to make the correct decision with regards to design and this encompasses the system as well as the part itself which is being considered. So it becomes a necessary point to make that one can now physically reproduce to microns in accuracy and no longer is it necessary to stick to statistical methods of assessment and aspiration. Physical biology can now be measured at a micron level as can the performance of these structures.
At a foundation has been a determined effort to move towards direct data transfer, from microscope image to layered manufacture, as it is called now. Because scaling effects exist, the non-Newtonian mechanical properties of the vast majority of hooked attachment mechanisms can only be mimicked and tested when manufactured at the same order of size.
The door is creaking open, upon the region of science and manufacturing technology called Microdesign. As never before the opportunity arises for manufacturing expansion into the realm of micron-sized structural designs that could benefit man through their use of their size. In the light of new developments into biomedical structures there is a need for stable materials at this scale to be used within biological systems.
The hook, as a shape of low-complexity, proved an excellent example to demonstrate the limits of current technology and its new abilities due to the work of Hirt et al. In terms of 3-D data collection via laser scanning, resolution of an overhang is impossible in C++ programming terms unless one moves the head of the layered manufacturing device in which case complex shapes can be reproduced. Surface modelling via Canny Edge Detection methods does not provide for holes or overhangs in the first instance.
The set of all Biological hooks in Nature can be divided along lines of material, structure and function. When considering shape and form one must consider it surprising that all biomaterials seem able to form hook shapes and do. At the smallest scale, near atomic level and in the region where self-assembly occurs, there must be incentive to form these shapes which is a directed response to the environment. It could be that these early shapes, these hooks, were in fact invented by Life itself as a form of camouflage with dual purpose and thereby were able to be used to vary Life without threatening it. For the first, the very first curve or hook shapes on earth must have occurred in the rock material of the surface and other parts.
A crude mapping system is available to us at any time, much like a parts manufacturer would catalogue a system of related parts. But this is not the purpose of the research, which is into micro-design of which the hook-shape forms a complex challenge.
Figure 5: The design space of attachment mechanisms. Micro-attachment mechanisms must find a space here. 
Figure 5 shows a design space for fasteners, without microfasteners included except in the form of gecko-feet.
1. Saunders B E, Biomimetic study of natural attachment mechanisms-imaging cellulose and chitin part 2. J. Robot. Biomim. 2015;2:7. doi:10.1186/s40638-015-0032-9.
2. Saunders B E, A biomimetic study of natural attachment mechanisms – Arctium minus part 1. J. Robot. Biomim. 2015:2:4. DOI10.1186/s40638-015-0028-5
3. Saunders B E, Microdesign using frictional, hooked, attachment mechanisms: a biomimetic study of natural attachment mechanisms – part 3. J. Robot. Biomim. 2016:3:4. DOI10.1186/s40638-016-0040-
4. Nicklaus, K. J. Plant, Biomechanics – An engineering approach to plant form and function (Chapter 10), Biomechanics and Plant Evolution, University of Chicago Press, (1992) , pp. 474–530
5. Gorb SN, Beutel RG, Gorb EV, Jiao Y, Kastner V, Niederegger S, Popov VL, Scherge M, Schwarz U, Vötsch W. Structural design and biomechanics of friction-based releasable attachment devices in insects. Integr Comp Biol. 2002 Dec;42(6):1127-39. doi: 10.1093/icb/42.6.1127
6. Hirt L, Ihle S, Pan Z, Dorwling-Carter L, Reiser A, Wheeler JM, Spolenak R, Vörös J, Zambelli T. Template-free 3D microprinting of metals using a force-controlled nanopipette for layer-by-layer electrodeposition. Adv Mater. 2016;. DOI:10.1002/adma.201504967.
Philos Trans R Soc Lond B Biol Sci. 2015 Feb 5;370(1661):20140027. doi: 10.1098/rstb.2014.0027.
8. Gorb E, Gorb SN Contact separation force of the fruit burrs in four plant species adapted to dispersal by mechanical interlocking. Plant Physiol Biochem. 2002;40:373–81
9. “Biological Mechanisms of Attachment, The Comparative Morphology and Bioengineering of Organs for Linkage, Suction and Adhesion”, W Nachtigall, 1974 translated by M A Biederman-Thorson, Springer-Verlag, ISBN 3-540-06550-4
11. Hasegawa M, Yoon S, b Guillonneau G, Zhan Y, Frantz C, Niederberger C, Weidenkaff A, Michlerad J, Philippead L, The electrodeposition of FeCrNi stainless steel: microstructural changes induced by anode reactions Phys. Chem. Chem. Phys., 2014,16, 26375-26384 DOI: 10.1039/C4CP03744H
COPYRIGHT BRUCE E SAUNDERS 2020 NOT TO REPRODUCED IN ANY FORM WITHOUT EXPRESS PERMISSION FROM THE AUTHOR OR PROFESSOR ANDREW PARKER, OXFORD UNIVERSITY
Copyright B E Saunders (2016)