Assisted 3d bioprinting

Abstract

A bioprinting apparatus 100 for 3D printing a model comprises a printing nozzle 115 for dispensing a magnetic or magnetisable bioink for forming the model; a bath 130 containing a model-supporting flu

Description

Assisted 3D bioprintinq Field of the invention The present invention relates to deposition 3D printing, whereby models or constructs are formed by depositing material in small amounts until the desired shape of the model or construct is achieved. Embodiments of the invention describe depositing sequential horizontal layers to form the construct but other deposition formats are possible within the scope of this invention. The invention relates 3D printing in the bioprinting field, particularly, but not exclusively to printing with bioinks, which are biocompatible printable inks used for example for tissue engineering and in biological research. It will be appreciated that it is preferable to utilise the invention in an enclosed environment where bioburden and temperature can be controlled. Background 3D printing is now an accepted manufacturing technique, usually adopted for complex parts and / or low volume production. Deposition 3D printing, also known as extrusion printing, or additive manufacturing, usually involves liquefying a material and laying it layer by layer on a substrate. The material sets or cures after deposition ready to accept the next layer. However, some materials cannot be treated in that way, for example bioinks which do not respond well to being heated or being exposed to strong UV light to cure them. In some instances, bioinks which contain organisms such as cells, cannot be heated to any useful extent. Printing of these ‘problem’ materials has been address by a previous patent application of the Applicant’s, namely published application WO2022 / 258938 with the title Additive manufacturing using low viscosity materials, the contents of which are incorporated herein by reference, and which describes a dual-material printing technique, where one material is used essentially as a mould, for a lower viscosity material. That technique is a useful way to hold low viscosity material together until, for example, cells have time to aggregate together, thereby allowing the mould material to be removed, for example using warm water. However, that technique does take time because two materials are used during printing. If the low viscosity material could be held together without a mould, at least until sufficient cohesion has taken place, i.e. aggregation, or setting of the material takes place, or some cross-linking of the material occurs, then the printing speed could be increased. The inventors of the present invention realised that some way of suspending ‘problem’ materials, (i.e. low viscosity materials or materials that slump once printed) is needed. US20210394391 discloses a thixotropic gel support bath to support a pre-ceramic material prior to curing. However, such thixotropic gels are not easy to remove from a delicate bioink printed construct once printing has finished. Hua et al, in “Fluid Bath-Assisted 3D Printed for Biomedical application: Form Pre Postprinting Stages” published in ACS Biomaterials Science and Engineering 2021 7(10), 4735-4756, provides a review of fluid bath assisted 3D printing using biomaterials. The paper sets out the problems associated with fluid bath assisted 3D printing in some detail. In particular, Hua et al mention that the rheological properties of the fluid bath can affect the speed of printing because, if the support fluid does not have good flowability, then a needle, or the like, used to extrude the printed material will drag in the fluid and the fluid will not close up behind the needle during printing, leading to a poor print. Also, the movement of the printing needle can agitate a viscous fluid bath and the already printed material. On the other hand, if the support fluid lacks viscosity, it will lack support for the printed material, and there is a tendency for the printed material to break into droplets, float, or sink. One technique discussed by Hue et al is the use of a thixotropic support fluid containing microbeads (a Carbopol granular microgel). The inventors have found that one problem with fluid baths is that they must be a reasonably specific density to match the density of the material being printed. Otherwise there is a tendency for the material being printed to float or sink, according to the density mis-match. This is a particular problem if more than one material is printed in the same construct, for example if a cell scaffold type material is printed and then a cell-containing material is to be printed. Also, the inventors realised that printing low viscosity bioinks becomes a challenge due to the inherent surface tension of the bioink that seeks to clump / pool together post print. The inventors have devised a technique for use when using a fluid bath to suspend a construct formed from one or more printed materials until it(they) become(s) a cohesive material e.g. it (they) become(s) aggregated, set, or cross linked, and able to support itself (themselves) out of the support liquid. That technique essentially employs magnetism to influence printed material, to aid the holding and / or positioning of the construct formed by that material, relative to a printer head at which the printed material is extruded into a fluid bath. Control of the positioning magnetic field allows the construct to be held / positioned accurately relative to the printer head, and the field can hold the print material substantially stationary relative to the printer head while more print material is added to the construct being printed. This positioning can be holding the construct stationary by applying a suitable magnetic attraction / repulsion force, or additional magnetic force to induce movement of the printed construct e.g. vertical and / or horizontal movement. The forces that need to be reacted during printing will be positive or negative buoyancy in the supporting fluid bath, as well as a drag from the printer head as printing progresses and material is extruded onto the already partially printed construct. The printed material is preferably one which is readily magnetisable, such as a ferromagnetically labelled material but, with a suitably strong magnetic field, less readily magnetisable materials may be employed, such as diamagnetic, paramagnetic and antiferromagnetic printable materials. The holding / correcting magnetic field used may come from permanent magnets, such as magnetic arrays, or from electromagnets, such a coil or coils of a conductive wire around a soft iron or ferrite core. The strength and uniformity of the magnetism (permanent or electromagnets) may be enhanced by the use of specifically directed pole arrays, such as Halbach arrays. Drawings The invention is set out in more detail in the examples described below, with the aid of the drawings herein, where: - Figure 1 is a schematic diagram of 3D printer apparatus according to one aspect of the invention; Figure 2 shows an example of a magnetic positioning element used in the 3D printer of Figure 1; Figure 3 is an illustration of a repulsive magnetic field; Fig 4 is a cross section of parts of the apparatus shown in Figure 1; Figures 5a and 5b are explanatory diagrams of a 3D printing process employing magnetic positioning; and Figure 6 is an alternative arrangement of apparatus. Detailed description Example 1 apparatus Herein the term ‘3D Microgravity printing’ is the printing of 3D models or constructs of magnetic or magnetisable printable materials, suspended during printing in a fluid. Referring to Figure 1, 3D bioprinting apparatus 100 comprises a print head 110 moveable in the illustrated X-Y plane, under the control of a controller 180. The print head 110 includes a print nozzle 115, which in this case is a fine needle with an internal exit diameter ranging between 0.05mm to 0.5mm, which reaches into fluid bath 130. The print head 110 further includes a nozzle withdrawal aperture 118, and a printed surface scanning device 112 / 114, which are described in more detail below. The bath 130 in use contains a print support fluid 135, typically a fluid that promotes crosslinking. The bath rests on a non-magnetic bed 120, in this case formed from aluminium. The apparatus further includes upper and lower magnetic rings 140 and 145 respectively. These rings act to magnetise material 150 printed by the needle 115 in use at an initial print height, shown as plane 155 in Figure 1, and hold the same substantially in place during printing. In this example, the flowable fluid 135 has a higher density than the printed material 150, and thereby the printed material 150 is exposed to a very low / zero gravity field due to material’s inherent positive buoyancy being cancelled out by the magnetic influence of a magnetic field of the rings 140 and 145 substantially opposing that buoyancy. The magnetic rings 140 and 145 are independently adjustable in the illustrated Z direction, for example by employing a stepper motor 170 and a cooperating lift mechanism 175, in this case a rack and pinion mechanism for each ring 140 and 145, controlled by controller 180. It can be seen that only one of the mechanisms 170 / 175 is shown in detail in Figure 1, but the same mechanism can be used for both rings. The mechanisms employed provide independent height adjustment in the Z direction for each ring relative to the print plane 155. A printed droplet Z position sensor, in this case a low power laser array 160 and a complimentary laser light detector array 165, is used to determine the position of the printed material 150 in the Z plane. That sensor is also able to confirm the position of the tip of the printing nozzle 115. Height position sensor data is communicated to the controller 180. Referring additionally to Figure 2, in this example the magnetic field of the rings 140 / 145 is generated by permanent magnets 142, where, in each ring the magnets are arranged as Halbach arrays. The magnets are held in a support frame 148, for example a cast resin ring. One ring 140 is positioned above a plane at which the printed material is deposited, and one ring 145 is positioned below that plane, each ring being arranged outside the fluid bath 130. The Halbach arrays consist of 10mm square N42 Neodymium magnets placed within a circular arrangement as depicted in Figure 2, with their polarities as shown and where polarities illustrated as wholly North [N], or South [S] have their opposite polarity on the opposite side to the side shown in the drawing. This configuration produces a magnetic field generated at 90° to the face of the circular array of Neodymium magnets and biased such that the inner part of the ring has a denser magnetic field than the outside. Thus, in the apparatus of Figure 1 the lines of magnetic flux propagate substantially vertically down from the top Halbach array 140 and an opposing magnetic field is generated vertically up from the bottom array 145, by flipping the bottom array 180 degrees relative to the top array, so that the respective patterns of the array are opposing each other i.e. North faces North, South faces South etc.. The inner diameter of the circular arrangement of Halbach array of permanent magnets is about 50mm which provides for uniform magnetic fields of a footprint size that encompasses X and Y coordinate movement through the magnetic fields. The orientation arrangement of magnet poles repeats for every 4 magnets such that a larger circular Halbach array can be realised by increasing the number of magnets by multiples of 4 again arranged as in Figure 2. Further these magnets are permanent magnets contained within a circular holding frame. Figure 3 shows schematically the magnetic repulsion of two adjacent magnets with similar poles, and the opposing poles of two magnets close to each other will produce a predictable magnetic field, such as the field generated in the arrays 140 and 145 Figure 4 shows a sectional view of the some of the apparatus shown in Figure 1. Additionally Figure 4 shows a schematic representation of the magnetic field F between the two rings 140 and 145, represented as dashed hatching lines. In practice the field will be more complex in orientation and strength. However, since the field is generally in opposing relation, there will be a very weak or zero strength field W in the middle of the field represented by the honeycomb lines. The field F and weak / zero field W will be largely unaffected by the support fluid 135 in the bath 130. The initial print plane 155 is below the zone W, so that there is a slight bias pulling magnetic print material 150, in this case downwards, against the positive buoyancy force on the printed material provided by the support fluid 135. Figure 5a shows the printed material 150 printed at the initial Z plane 155, and two layers printed on top. These additional layers are shown in the exploded view of Figure 5b, where printed droplets 152 can be seen having magnetic polarity indicated by shaded (North) and unshaded portions (South) all aligned and thereby held in place together under magnetic attraction. The attainment of the successive print layers-A to start, then B, then C etc, are explained below. Example 1 operation Figures 1 and 4 show the Halbach arrays 140 and 145 that are formed by a circular arrangement of magnets; the overall diameter of which dictates the magnetic field coverage and therefore the available X and Y printer head movement footprint. Under the influence of a conventional X-Y 3D printer mechanism controlling the movement of the printer head 110, the tip of the print nozzle 115, for example, a Lee Products Ltd 0.05mm resolution printer head nozzle Part No INZA4620928T enters the bath 130 from above to the virtual print bed Z = 0.0 axis position 155 (as determined by material Droplet Optimisation Protocol position) through the centre of the top Halbach array 140 and moves in X and Y at a physically constant fixed height, initially corresponding to Z= 0.0 at start of print - first layer print- level 155 illustrated, all within the magnetic environment created by the Halbach arrays; as dictated by control software. This is achieved by utilising a 4D extrusion bioprinter dispensing bioink having magnetic properties; that employs opposing magnetic fields as described above, arranged above and below the Z axis position 155 of the printer head droplet deposition position. The opposing magnetic fields are generated by the two circular Halbach arrays of permanent magnets 140 and 145. Figure 2 shows the notional origin of the X-Y plane at the top right-hand corner. So, the print size is in this case confined by the inside diameter of the support frame 148. The available X and Y printer head movement footprint can be increased by increasing the number of magnets together with their spaced positions within a larger diameter circular holding frame 148. The invention is intended to be utilised in a temperature controlled clean room environment to employ opposing permanent magnetic plates (Halbach arrays with configuration as shown in Figure 2) situated in an enclosure shown schematically as enclosure 101 in Fig 1, to encompass the printer head and physical printer bed. The printer head 110, bed assembly 120 and fluid support bath 130 are preferably manufactured from non-magnetic components, for example aluminium, plastic or nonmagnetic stainless steel. The two opposing fields of the Halbach arrays 140 and 145 act on a magnetically labelled printed material 150, to magnetise the material in such a way that the material is biased toward the field generated by the lower Halbach array 145 due to the nozzle print position 155 being closer to the lower Halbach array 145. Given the Halbach arrays are both producing say a North pole magnetic field then the printed material will align itself such that its South pole is aligned downward as attracted by the stronger lower North pole of the lower Halbach array; and be held there along with all the other similarly magnetised printed droplets. That droplet orientation are positioning is illustrated in Figure 5b. Droplet Optimisation Protocol Initially the outlet pressure of a microfluidic pump (within head 110) is adjusted to determine the pressure that just extrudes a droplet of bioink and a corresponding lower pressure that just inhibits droplet extrusion. These two pressures will be close to each other in value and can be stored in memory for later use. Virtual Z-axis Bed calibration and stabilisation The initial start print positions of the Halbach arrays 140 and 145 are determined using test material droplets for a desired bioink that has been labelled with magnetic material e.g. Fe3O4. The bioink material to be printed in this example has a lower density than the complementary crosslinking material contained in the bath 130. The Halbach arrays are initially positioned at their minimum retracted positions (providing the smallest gap between the Halbach arrays) utilized to generate magnetic fields producing upward and downward forces on magnetically labelled bioink droplets dispensed from the printer head. Initially the position of the Halbach arrays will result in the bioink droplets settling at the position of the lower Halbach array by virtue of its strong magnetic attraction overcoming the bioink buoyancy. The position of the Halbach arrays are then successively adjusted to widen the gap between the Halbach arrays; such that, in a first case where the Halbach arrays are populated by electromagnets which are depowered during movement and powered when stationary at their new incrementally widened positions, then eventually subsequent bioink droplets 150 are positioned according to the equilibrium of the lower Halbach array attractive magnetic force balancing the force of droplet buoyancy. In a second case, where the Halbach arrays are populated by permanent magnets and their positions are incrementally widened; then subsequent bioink droplets will be repeatedly dispensed at each of the widened Halbach arrays test positions; in order to realise / determine the equilibrium 150 droplet position described above. The final Z height of the droplets 150 is detected by the laser detection system 160 / 165 shown in Figure 1. This process effectively calibrates the position of the Halbach arrays to hold droplets 150 of the bioink at the Z = 0.0 axis printer head nozzle position (print first A layer start Z -axis, position Figure 5b). From this Z = 0.0 axis position the Halbach arrays are further slowly (to maintain droplet capture / tracking by the lower magnetic attracting Halbach array) widened until the laser detection system 160 no longer detects the print droplets 150; this provides a calibration of delta Halbach arrays movement to realise a 0.05mm movement of the bioink droplet. Once these Halbach array positions have been determined a virtual printer bed is realised and corresponds at the physically configured print head nozzle position where Z = 0.0 (first print layer - print start Z position); together with Halbach array delta widening positions that provides for a 0.05mm Z axis downward displacement of a droplet 150 of the bioink, so that successive layers B, and C (as illustrated in Figure 5b) can be printed at the plane 155. Since the printed plane material 150 conserves its magnetism, the printed material will be reasonably positionally stable. However, it is envisaged that minor adjustments in the positions of the magnetic rings 140 and 145 may be required, and for some print materials that do not conserve their magnetism very well, continual adjustments will be required, which will be, essentially, continually recalibrating the virtual Z axis 155 as described above. The printer head is physically configured to always print at the same z position (height) throughout the print process. The printer head will perform movement perpendicular to the two Halbach array arrays. This is achieved by fixing the printer head to a conventional X-Y movement mechanism that can move the printer head 110 in a 2D X / Y manner within the microgravity void generated by the Halbach arrays, all under the control of suitable software. The layer print is achieved by the extrusion of droplets of bioink material at the required X / Y coordinates to construct the model layer. Unlike conventional 3D printing, since the printed material is supported in a fluid bath and held in place in a magnetic field, it is not necessary to have a printed layer under successive layers, because the material is capable of bridging gaps in the underlying print. For printing of additional / next model layers (B,C etc) the virtual printer bed is moved down (model Z axis increasing) by adjusting the gap (widening) between the Halbach arrays. These Halbach array adjustments cause the magnetically captured droplets printed in the previous layer(s) to move down in height from their previously printed positions due to their attraction to the now moved lower Halbach array (the magnetic fields are uniformly produced by the Halbach arrays with their magnetic forces uniformly acting on each printed / suspended Bioink droplet). These Halbach array adjustments can essentially move the virtual printer bed downwards to effectively increase the Z print height by the required print resolution 0.05mm. The next layer to be printed is achieved by again performing the extrusion of droplets at the physically fixed print head height Z axis position; whereby the dispensed droplets will be attracted to the droplets held at the previously printed model layer. The droplets stationed at the previous model printed layer now act as supporting droplets / constructs for the next layer droplets. This is similar to how an FDM printer operates (employing supporting constructs) except that if the dispensed droplet does not have a nearby droplet to attract to on the previously printed layer; then the droplet just extruded will escape the magnetic attraction of the Lower Halbach array and due to its low density buoyancy and subsequent magnetic attraction realignment with the Higher Halbach array will float upwards towards the Higher Halbach array and / or surface of the crosslinking material contained in the bath. In a refinement, it is possible to employ the technique similar to that described in the aforementioned patent application WO2022 / 258938, such that a sacrificial bioink, say magnetically labelled gelatin is printed at some or all of the layers (A,B,C etc) to provide support for a subsequent less cohesive construct / model bioink deposition. For example; once a model bioink has been printed at the model layer; the inverse of that model is then printed at the same layer with the sacrificial bioink, surrounding the model, resulting in the two bioink depositions providing a flat surface which also serves to extend the magnetic attraction field of the lower Halbach array to enhance the magnetic field for the next Z axis model layer print (Figure 5). For this approach both the model and sacrificial bioinks must have similar density. In order to alter their density water can be added to or removed from some hydrogels (especially gelatin). In the example above, the selection of the support fluid in the bath 130 must have a higher density compared to the bioink(s) to be printed. During printing the printer head nozzle 115 moves in the 2D X / Y plane and dispenses material in a conventional manner, at a dispensing pressure as determined above. The dispensed material droplets will remain suspended in the support fluid 135, at their final resting deposition positions as dictated by the virtual printer bed (Z plane 155). To negate disturbance of the printed model layer(s) during layer printing the printer head nozzle 115 is preferably moved very slowly 6mm per minute to print the model layer in a raster of model slices through the Y axis; utilising alternating right to left then left to right X axis model slice starting positions for printing successive slices. In addition, the movement (for example widening) of the Halbach array configurations (to increase model print Z height) for subsequent model layer prints are also moved slowly at 6mm per minute; again to negate disturbances of the printed model. The extrusion of bioink droplets is performed using microfluidics and two pump pressures determined during the Droplet Optimisation Protocol activity. The two determined pressures; one to achieve droplet extrusion; and the other to inhibit droplet extrusion, are very close to each other by their very nature and this serves to minimise, when material dispensing, any disturbance of the fluid contained in the bath 130 and therefore any disturbance to the already printed model artifacts. Since the bioink droplets 152 are suspended in fluid with complementary magnetic forces acting in concert this arrangement will inhibit surface tension pooling / clumping of the droplets and thereby maintain the model structure enabling crosslinking to take place. This arrangement also has the benefit that no time must be expended to wait for lower layers to crosslink prior to printing further layers on top. Print Accuracy Verification Protocol In recognition of the possibility that the magnetic fields F and W mentioned above may not hold the printed materials in an exactly the desired position, for example in the Z plane, the hardware of Figures 1,5 and 6 have the laser scanning device 112 / 114 mentioned above to check for flatness, and allow compensatory adjustment if needed. That scanning device is intended to work in conjunction with a needle withdrawal mechanism, in this case a stepper motor 116 which lifts the needle 115 upwardly, through aperture 118 as illustrated in Figure 5. With further reference to Figure 5, the laser scanner 112 / 114 provides a fine beam of laser light L from a laser 112, which propagates toward the printed layer 155, and is partially reflected back to a laser light detector array 114. Triangulation of the reflected light can provide a measurement of the height in Z of the printed layer 155. A correction in Z can then be made at the next print attempt, with all other parameters remaining the same. For a more refined print accuracy verification, errors in X, Y and Z can be determined with the hardware above. The Applicant’s previous published patent application WO2023 / 218159 with the title GRAPE DATA FORMAT AND METHOD OF 3D PRINTING, the contents of which is incorporated herein by reference, discloses a 3D printing data format which reduces the amount of data needed to define a 3D print (i.e. a 3D print file), by defining each print layer as a series of artifacts with coordinates in X,Y and Z, in this case rectangular artifacts. To facilitate the XYZ accuracy verification technique mentioned immediately above the GRAPE formatted data described previously can be enhanced with an additional 3 displacement components XYZ defined for each rectangle which represent any deviation from where the print should be. The scanning device 112 / 114 can be moved in X and Y over the whole print bed, including the printed layer 155 to provide a Z height feedback to the control software for each printed rectangle, and further enhanced by recording the Z height error- plus or minus the expected Z height determined by the scanner, for example into the print file. In addition, the scanner will detect material around the edge of the printed layer 155 and determine if the material is correctly located in X and Y by finding the true edge of the printed material and using control software running in the controller to determine if the true edge coincides with the intended print edge. The scanner will detect using the reflected laser light one of:-material where material should be, material where no material should be, no material where no material should be, or no material where material should be. In each case positional correctness or error can be determined. The control software, on the print layer being verified by the scanner will decode the scanner output together with construct’s GRAPE format data as described in the previous patent application, to determine actual XYZ positions of the 4 corners of the rectangular artifact and seek to make adjustments to these points by accumulating and applying positional displacements (based on the Z height error and / or any deviation in the X and Y coordinates of the corners of the rectangles) to append to the data defined for the construct(s) in the GRAPE data format. An example of an unenhanced GRAPE data format is given the table below, where a rectangular construct to be printed is defined by its corner X&Y coordinates and a common Z coordinate (i.e. the height / thickness of the rectangular construct): Table 1 1st corner X&Y coordinates 2ndcorner X&Y coordinates 3rd corner X&Y coordinates 4th corner X&Y coordinates Common Z coordinate 8.0000e+01 6.4950e+01 8.0000e+01 6.5050e+01 8.5600e+01 6.5050e+01 8.5600e+01 6.4950e+01 1.0000e+01 In the enhancement mentioned above the GRAPE data format would be modified to apply more data which data can be used to modify the printer head path during subsequent printing (for example a new print attempt, or a new print layer) to take account of printing errors in the prior or subsequent layer(s). For example the modified data may be recoded according to the table shown below: Table 2 1st , 2nd, 3th and 4th corner X&Y coordinates as tabulated above in table 1 Displacement required in X Displacement required in Y Displacement required in Z 8.0000e+01 6.4950e+01 etc 0.0000e+00 0.0000e+00 0.0000e+00 In the second table it can be seen that a positional displacement is added to provide an error correction, which in turn was found by scanning the printed material. In the example above the correction displacement is zero, indicating no errors where found. In addition to providing displacement error adjustment; this displacement data can also be employed on successive model rectangular constructs to provide overall model rotation in the XYZ planes. Print Verification Operation In practice the laser displacement sensor 112 / 114 on the printer head 110 is capable of measuring a single point on the printed model using triangular reflection. The steps for error determination are:- 1. Model first layer is printed in the usual manner in the X and Y plane at the Z virtual printer bed height 155. 2. Nozzle is extracted from the cylinder containing the cross linking material (CLM) so that the laser displacement sensor (LDS) can be operated to perform an area scan of the model (again the LDS is outside / above the cylinder containing the CLM) 3. LDS (as moved by the print mechanism in the X and Y planes) scans the complete available printable footprint area to detect model droplets 150. LDS performs this task using a raster scan technique for all available points based on the printer movement XY resolution in this case 0.01mm (e.g. for movement resolution 0.01mm for first line xO.OO y0.00, x0.01 y0.00, x0.02 y0.00...and so on for the complete printable x range (0.00 to max x) and similarly the raster scan repeated in the y direction (xO.OO y0.01, x0.01 y0.01.. etc) and repeated for the complete printable y range (0.00 to max y). The LDS as well as detecting the presence of the droplet at this xy point position also provides distance of this droplet from the LDS . This provides for a measurement of delta z. The LDS derived data from the scan of the printable area is communicated to the controller. 4. steps 1 to 3 repeated for all model layers. 5. Controller uses the feedback LDS data to determine model displacement xyz to be applied to each printed model rectangular construct(s) to facilitate adjustments of the printer nozzle positions for a subsequent print of the model., with the goal of negating any printing positional errors that occur say due to uniformity variations of the magnetic fields F and W. Since it is likely that any errors in print position will be repeatable between prints, it is possible to store any displacement components derived from the scans of earlier print attempts and use those displacement component values to modify the print path of the subsequent print attempt, in order to pre-empt any print errors. Once the model print has completed, then the print head nozzle 115 is once more extracted from the bath 130. Crosslinking of the model occurs generally without further intervention, but the bath temperature can be adjusted to increase the rate of crosslinking, and for some bioinks, additional light can be used to increase crosslinking. Once crosslinking is complete, the rings 140 / 145 or 240 / 245 can be raised whilst maintaining a suitable gap between them. This gently raises the model to the surface of the fluid 135 in the bath 130, so it can be removed as a completed and self-supporting construct. Magnetic influence mapping In another embodiment of print verification (printer self-calibration) reguiring no error displacement data to be included in the model GRAPE data file, a method for initial calibration of the magnetic influence has the following steps: a) The control software reads the model GRAPE data file to determine the minima and maxima of X,Y, and Z coordinates of the model to be printed, b) Employing this data, the control software prints a complete sguare or other regular shape layer using the print head resolution at virtual printer position 0.0 based on the minimum and maximum X Y coordinates encompassing the extremities of the model to be printed. c) This now printed layer is then scanned in the XY and Z directions for flatness and accuracy of material deposition in comparison to the defined sguare (or other regular shape) coordinates with any errors in the material deposition being recorded by the control software, for the respective Z height position. d) The control software then adjusts the magnet(s) to move the layer down by the Z movement resolution. e) Steps c and d are repeated until the maximum model Z height is achieved. f) The magnetic influence is adjusted to allow the printed calibration sguare (etc) layer to rise to the surface of the bath for removal. g) The final step is proceeding with the model print; applying the error corrections determined at each XYZ print space position as recorded by the control software during the calibration steps described above. Materials and applications Embedded metallic biomaterials may be used in the printable bioink. Their use in biomedical devices and components, such as hard tissue replacements, cardiac and cardiovascular stents, etc. demonstrates that it is possible to use such materials in bioink and utilise their magnetic properties. In particular, metal ion cross-linkable hydrogels are suitable for use in this application because they have excellent self-healing, fast recovery, biocompatibility and high mechanical strength and have good multi-stimuli responsiveness. One example of such a suitable hydrogels is iron oxide Fe3O4 nanoparticle-loaded alginate (ALG) hydrogel. However, most readily printable bioink materials can be employed with the invention described above, for example materials containing collagen types 1 to 28, jellyfish collagen, nascent protein polypeptides, deoxyribonucleic acids (DNA) or ribonucleic acids (RNA), or their combinations. Other, potentially sacrificial, materials could be used also such as one or more of gelatins, other alginates, or thermo-responsive hydrogels. The printable materials mentioned immediately above will be seeded or labelled with a magnetic component, for example a ferromagnetic component in the form of Fe3O4 beads. With a suitably strong magnetic field, less readily magnetisable materials may be employed, such as diamagnetic materials e.g. copper or carbon, or paramagnetic materials e.g. aluminium or oxygen, or antiferromagnetic printable materials e.g. chromium. The support bath fluids mentioned could be any fluid such a water. However, a preferred fluid would be a solution containing a bioconjugation reagent (cross linking agent of which there are many) of a type suited to the material being printed. For example where collagen is printed then a biotin reagent could be used. Avidins, photoreactive crosslinkers such as , Phenylaz-based reagents or Benzophenones could be used. Other reagents are commercially available. The printed material may also contain cells for example eukaryotic cells, and the support fluid may contain nutriments for those cells. Example 2 Apparatus Figure 6 shows an alternative arrangement of the bioprinting apparatus described above. In this alternative, most of the apparatus is the same and functions as described above, and in particular the scanning arrangement 112 / 114 is maintained. However, the magnetic rings 140 / 145 have been removed and replaced by a equivalent electrically conductive coils 240 and 245, wound circumferentially around the bath 130. In this example the coils are about 20 turns of fine enamelled copper wire 242 / 247 of about 30 swg. Example 2 operation Current is caused to flow by potential difference in both coils with the current flowing in opposite direction. Once current is flowing in the coils opposing magnetic fields will be generated in the middle of the coil with lines of flux approximately coaxial with the coils’ axes . In turn those fields can be used in a similar manner to that described above to influence the magnetisable bioink, stopping it floating, and the multitude of printed droplets 152 will form a cohesive mass held together initially by the influence of the magnetism. Whilst it is possible to use the Z height adjustment 170 / 175 as descried above, as an alternative or to compliment that physical height adjustment, it is possible to adjust the current in the coils by means of the controller 180 shown in Figure 1, for example in an analogue manner, or by pulse width modulation to alter the strength of magnetic field in order to lower the printed material in the bath, and print another layer on top, as described above. In this case the magnetic field strength of the lower coil is temporarily increased to attract the droplets to the desired z position and subsequently the magnetic field strength lowered to maintain droplet buoyancy equilibrium at this new Z position. The desired Z droplet position being verified by the control software using the Z laser detector device. In this example, alternating the coil current allows the printed material to be demagnetised, for example after cross linking of the model has finished. It will be appreciated that the numerous features described above and / or illustrated herein are set forth by way of example and are not intended to limit the scope of the invention. Numerous alternatives, variations, modifications, additions, and omissions, to those examples will be apparent to a skilled person in the relevant art. Halbach arrays have been described in the first example, although these could be replaced by individual electromagnets with equal utility. Circular components have been illustrated, but other shapes, for example square, triangular, hexagonal or other regular or irregular polygons could be employed. X,Y and Z cartesian coordinates have been described, but the skilled person would readily understand that other manyfold coordinate systems could be use, for example coordinates characterised by angles or polar type systems. It is envisaged that features from different embodiments may be brought together, without adding to the scope of the invention. In addition, the order of any features in the form of method steps or sequences in the description, claims and / or drawings herein is not intended to require that order of performance unless a particular order is necessary for technical reasons. Multiple features in a single claim herein may be so combined in that claim for, for example, fiscal, not technical reasons and so such combined features are not necessarily intended to form a whole inseparable technical concept. Thereby, in the claims set forth, it is intended that claim features may be added to, exchanged between, or extracted from, claims containing other features without broadening the scope of the invention defined by those affected claims, or causing a so-called intermediate generalisation.

Images