Latex-based aqueous UV-curable inks

EP 2 960 306 A1
Aqueous radiation curable inkjet inks
Agfa Graphics NV

UV curable ink jet inks lacking solvents have been used extensively in industrial applications, because of their reliability and because they adhere to a multitude of substrates. One disadvantage of these “100%” solids UV curable ink jet inks is that relatively thick layers are printed, compared to aqueous ink jet inks which generally contain less than 20% – and usually less than 10% – solids. Thin layers generally have better flexibility than thick layers and so a low solids UV curable formulation would potentially be advantageous. However in most other aspects, low solids aqueous latex ink jet inks tend to exhibit an inferior performance compared to UV curable inkjet inks.

A latex is a stable colloidal dispersion of natural or synthetic polymer particles in an aqueous medium. Synthetic latexes can be made by polymerising monomers, generally (meth)acrylates, that have been emulsified with surfactants, and such materials have been frequently used in ink jet inks. Acrylate based polymeric particles often have a relatively high minimum film formation temperature (MFT), which prevents the use of temperature sensitive substrates for latex ink jet inks. The MFT tends to be very closely related to the glass transition temperature (Tg) of the latex polymer and so thermosensitive substrates could feasibly be used if the Tg of the latex ink was low, below 0C for example. Unfortunately, such low Tg latexes tend to show poor jetting properties and blocked nozzles as the latex can easily film form in the nozzle.

This patent suggests that low MFT latex dispersions can be jetted successfully if fully water insoluble (meth)acrylate monomers are mixed into this latex. It is believed that this liquid polymerisable compound rapidly migrates into the surface of the hydrophobic polymer particles, forming a barrier and protecting them against agglomerating and fusing together. This is suggested to be particularly advantageous for low MFT (<0C) polyurethane particles which, without the exemplified monomers, can only be reliably used in ink jet inks at very low levels when high levels (>10 wt%) would need to be used to confer the required physical properties to the final printed film.

The experimental section discusses this approach in detail. One example took 100g of a 42 wt% solids polyurethane dispersion (Bayhydrol UV XP 2689) and added this to 100g of water. To this was added 5.3g of the monomer dipropyleneglycol diacrylate (DPPA) and stirred for 20 minutes at 800 rpm using a Disperlux mixer. This mixture was then filtered and used in an ink jet ink at 23-35 wt% along with 20 wt% of a colour pigment dispersion (Diamond D75C – a 15 wt% dispersion of C.I. Pigment Blue 15:3 in water), humectant (10 wt% 2-pyrrolidone and 20 wt% 1,2-hexane diol), photo initiator (1.1 wt% Irgacure 500) and water. Comparative inks were also formulated using only the polyurethane dispersion without the addition of the UV curable monomer or photo initiator.

These inks were evaluated on a Dimatix DMP2831 ink jet printer (10pl head) with a cartridge temperature set to 24C, at a firing frequency of 5 kHz and a firing voltage of 20-25V. The results show that at the higher levels of polyurethane dispersion, only the exemplified monomermodified dispersions show good ink jet performance, whereas the unmodified inks would not jet.

Further inks were formulated using this dispersion mix with an aim to investigate various photo initiators. Three groups of photo initiator were tested, water soluble, water insoluble and acrylated photo initiators. These inks were tested by applying them on polyester film, drying at 60C for 3 minutes then curing using a Fusion VPS/1600 lamp at 20 m/min. HPLC was used to test for the presence of any extractable photo initiator from the cure films. The results clearly show that the reactive acrylated photo initiators resulted in far less PI leaching from the cured film. No data is presented regarding mechanical properties.

Decorating on coffee

US 2015/0251470 A1
Method and apparatus for applying designs on the surface of a beverage
Steam CC Ltd.

This patent application from an Israeli company is a machine for printing onto frothed milk on the top surface of premium coffees, such as café au lait, cappuccino, latte or the like. It is already carried out manually, by pouring coffee solution onto the frothed milk in a pattern, pouring frothed milk on the coffee while moving it, or by using a utensil dipped in coffee to trace a design.

Sept-15The printer consists of an ink jet printhead on an X-Y gantry so that the printhead can be moved across the top of the coffee cup 9 in both axes. The cup isn’t moved so as not to disturb the foamed milk and hence the image being printed. The only adjustment involving the cup is the height and hence spacing between the foamed milk surface and the printhead, adjusted by table 8.

The “ink” printed is a coffee solution. The system enables much more elaborate designs to be printed compared to manual methods, plus repeatability.

Ink jet dye discharge printing for fabrics

WO 2015/025310 A1
Dye discharge inkjet ink compositions
Kornit Digital Ltd.

Dye discharge printing is a method of forming a design on a dyed fabric by printing onto this surface a colour-destroying (dye discharging) agent to create a white or light pattern. This discharge ink can also include a colorant that is unaffected by the discharge chemistry, thereby allowing light, vibrant colours to be printed onto darkly dyed backgrounds, with perfect registration.

An alternative approach to dye discharge printing, often used for printing onto dark coloured T-shirts, is one in which a thick opaque white background ink is first applied to the fabric before the subsequent colours. Although this is a cost effective and relatively simple approach, the resultant fabric can have a compromised feel due to the quantity of materials deposited on the surface. In comparison, dye discharge printing “removes” colorants resulting in a fabric with a soft hand-feel in the printed areas and so is often used for more luxurious fabrics.

A common discharge agent is zinc formaldehyde sulfoxylate (ZFS) that, although effective, has a short pot life of around a working day. Solutions of this material tend to degrade into insoluble side products of zinc oxide and zinc sulphide which can go on to block ink jet printheads. A discharge ink suitable for ink jet printheads would have to be stable for months, and ideally a year, to be useful. This patent suggests that this can be achieved with the ZFS discharge agent if it is formatted with an excess of a complexing agent such as ethylene diamine tetra acetic acid (EDTA ).

Specifically the ratio of the ZFS:EDTA should be less than 0.6:1 to achieve lifetimes of greater than 1 month and, in some cases, up to one year. This combination was formulated into an ink jet ink formulation with 7 wt% ZFS, 18.9 wt% EDTA , 37.2 wt% various humectants, 0.2 wt% surfactants, enough acetic acid to neutralise/buffer and water to make up. This was loaded into a “Kornit Avalanche” T-shirt printer and used to print onto a black T-shirt at various levels from 30 to 100%. As the black T-shirt used was not pre-bleached before dying, the base colour of the cotton is a light brown colour, requiring a small amount of a white ink to bring the colour towards pure white.

The exemplified test print demonstrates that a good white can be achieved on black T-shirts at significantly lower pigmented white ink levels.

Metallic effects from a thermal ink jet ink

US 2014/0170384 A1
Neutral gray reflective ink
Hewlett-Packard Development Company

Judging by the number of patents filed, it appears that the ink jet printing of expanded colours and appearances are sought after for home and office decorative printing. Reflective inks, in particular, have been disclosed with a particular emphasis on metal flakes – materials currently used in more traditional metal-effect printing processes. Thermal printheads have significant constraints regarding viscosity and particle size and so large metal flakes are a non-starter and a silver nano-particle approach would be expensive and questionable from a toxicological perspective.

The patent suggests that to look metallic, then only a certain threshold value for specular reflection needs to be achieved. It is suggested that 10% of the incident light must be specularly reflected, but greater than this is always desirable. With this specular reflectivity requirement relaxed, alternative non-metallic materials can be considered, such as magnetite. Magnetite (Fe3O4) can be made up into a stable nano-dispersion, suitable for thermal ink jet, and will give a metallic lustre when printed on glossy paper.

However, magnetite has an inherent yellow-orange hue and so to achieve a neutral silver effect a mixture of blue, cyan or magenta dyes must be added to counteract this. The exemplified ink jet formulations first prepared the magnetite dispersion in water by heavily bead milling a 5.6 wt% mixture of the oxide in water with a polyether alkoxysilane reactive dispersant at 0.5 wt%, relative to the pigment. This was then used (at 36.3 wt%) to make a simple thermal ink jet ink with 14 wt% humectant, 0.1 wt% neutralised styrene acrylic binder, 0.3 wt% surfactants, 5.4 wt% of a cyan and magenta dye mixture and water.

June-14aThis ink was loaded into a HP black cartridge from a Photosmart 8450 printer to print colour test patches. The chart left shows the colour coordinates of the neutral formulation above, compared against the non-colour corrected magnetite dispersion and a comparative silver nano-particle ink. It can be seen that the yellow tint has been fairly well counteracted in comparison to the metal ink samples.

The patent goes on to discuss the importance of the media in achieving a metallic effect and that the pore size of the surface of the media must be smaller than that of the magnetite pigment particles to ensure they remain on the surface as a contiguous layer. Fig 1 below shows this schematically with the dyes (16) absorbing into the media under the ceramic particles (14). There is an obvious issue to such a configuration in that the robustness of the printed image will be compromised if the pigment is left unprotected on the surface, but this is addressed in the patent application US 2014/0170395 A1.

June-14bAlthough it seems that the ultimate specular reflectivity of this approach is not quite that of a metallic silver ink, there do seem to be significant advantages in using such benign and low cost iron oxide materials.

Digital binder printing for décor applications

US 2014/0028772 A1 Digital binder printing
WO 2014/017972 A1 Digital binder printing
Floor Iptech AB

These patent applications describe a method of printing decorative boards and in particular flooring. The inventor is also CEO of Välinge, a leading R&D company in the flooring industry, and both companies are based in Viken, Sweden.

There have been several proposals to print onto laminated board or flooring using ink jet. Traditional printing is only cost efficient in very long runs. However the cost of ink jet printing laminate flooring is around €1/m2 assuming the cost of the ink jet ink used is €100/litre. This is 10 times the cost of ink for conventional printing. In addition ink jet inks are limited by viscosity and therefore the solids contents that can be used.


The solution is to use ink jet heads to form a wet image on the substrate, then apply a powder of resin and pigment, and vacuum off excess powder from the non-image areas. The ink jet head 32 jets a liquid on to the substrate 2 to form the desired image 11 of the rst colour. The liquid can be just water, as long as it remains wet enough for the next step. The substrate typically has a top coating of a thermosetting resin, preferably a melamine formaldehyde.

Next the powder is applied by scattering unit 27, consisting of an embossed drum and an oscillating brush. The powder is a mix of the first colour pigment and melamine formaldehyde. Where the image is wet the binder melts. This ‘image’ is then cured and dried by UV lamp 23. The powder in the non-image areas that was just scattered on the substrate surface is then removed by the vacuum unit 28. The powder that is removed is returned to the feed hopper for the scattering unit.

Other colours can be printed in exactly the same way until the image is complete. The board is then subjected to 40 bar pressure and 160C to cure and form a hard wear-resistant surface.

The cost of the powder reduces the cost of printing to around €0.2 m2, one fifth that of direct printing with ink jet.

Inside Landa’s Nanography process

WO 2013/132418 A2
Digital printing process
Landa Corporation Limited

This patent application describes the complete process and makes the basic claims for the overall process.  The first drawing is a schematic and is not to scale.  Multiple array printheads 300 print an image on to a heated intermediate substrate belt 130 which rotates clockwise around rollers 104 and 106.  The image dries partly due to the heated belt but with additional heating from above (not shown).  The printheads can be moved backwards from the belt to maintenance and capping stations.

Sheets are fed from a conventional offset press sheet feeder 506 and are transported through the machine by a series of gripper drums.  Drums 502 and 504 bring the substrate into contact with the image on the belt, transfer being enabled by pressure from back-up rollers 140 and 142.  After the first side of the sheet has been printed at the first impression, it is passed by a further gripper drum to the perfecting cylinder 524.  The perfecting cylinder has a circumference twice that of the other drums.  The grippers are timed so that as the trailing edge passes drum 526 the grippers grab the rear edge.  At the second impression stage the back side of the sheet is then printed.  In this way the single imaging engine can be used to print duplex.  To do this the images for Sep13-2
each side of the sheet are formed alternately and mirror-imaged so that after transfer they are correctly orientated.

After transfer to the sheet, the ink is dry enough to be handled by the gripper drums and the output stacker 508.  No further curing or drying of the image is required.

Other configurations are possible, and in fact a simplex web press was shown at Drupa 2012.  There is just one impression cylinder and pressure roller pair 502 and 140.  To ensure none of the web substrate is wasted, it’s fed intermittently past the impression cylinder to close up any gaps between the images on the transfer belt.  As the web drive has considerable momentum, the intermediate feed is achieved by using dancer rollers 554 and 556.

Going back to the duplex sheet-fed press, the figure below shows the intermediate substrate or blanket support system with the blanket and front support removed.   The roller 104 serves to pull the belt past the printheads at a constant speed, while roller 106 is also driven but is adjusted to keep the belt in a state of constant tension across the top run.  Roller 104 is mounted in pillow blocks so that it can be moved to allow for belt stretch.  Pressure rollers 140 and 142 are mounted inside the support, press against the blanket at the transfer points to the substrate, and can be raised and lowered.

The blanket runs in contact with thermally conductive support plates 130.  The junctions between these plates are deliberately staggered to prevent the creation of a line along the belt.  Heater elements 132 under the plates supply the heat to the belt along the upper run.

Not shown is a fan system to create a negative air pressure within the blanket support box.  This helps keep the blanket in contact with the heated plates, and also keeps the bottom run of the blanket from contacting the impression cylinders when the pressure rollers are withdrawn.

Each of the sides of the blanket support has a track 180 which engages in features fixed to the sides of the blanket.  This serves two purposes.  Firstly it enables the blanket to be threaded through the machine by entering it into the guides.  Although the blanket could be made as an endless belt, it is more economical to make it as a sheet and join the ends within the machine.  Secondly it steers the belt past the printheads and maintains tension across the belt.

To the left is a top view of belt 210.  Alongside each side are features 270.  These can be one half of a zip fastener that has been sewn to the edges of the blanket.  The projections 270 are contained within the channel of track 280.  Roller bearings 282 can be used to reduce the friction.  As the belt is heated the components must be capable of withstanding up to 250C.  Suitable materials for the projections include a polyamide reinforced with molybdenum disulphide.  The guides are used to stabilise and tension the belt as it passes the printheads to ensure accurate image formation, but can be omitted through the drying station and in other places where the belt is slack.

The blanket is formed into a continuous belt using a zip or hooks and loops fastener.  This may be supplemented by adhesive or tape.  It is important that the seam is the same thickness as the belt itself, otherwise there will be speed variations when the seam passes over the driven rollers.

Details of the ink can be found in Directions September/October 2013 page 17, and materials and structure of the blanket in the media reviews in this issue of Direction on page 22.  The surface of the blanket has a thin upper release layer that is hydrophobic, such as a silicone material.  This is formed on a reinforcement layer, such as a fabric.

A pre-treatment liquid can be applied to the blanket by a roller mounted opposite tension roller 106.  This coats the blanket with a thin dilute film of charged polymer to modify the surface properties.  By the time the blanket reaches the printheads the film is completely dry.

The purpose of this chemical treatment is to counteract the surface tension of the aqueous ink when it contacts the hydrophobic release layer of the blanket.  Without the pre-treatment the drops would flatten on the blanket surface on impact, and then surface tension forces would restore the ink into drops or beads.  With the pre-treatment the negatively charged polymer particles in the ink drop adjacent to the surface are attracted by the positively charged pre-treatment film.  This “freezes” the drop after impact so that beading does not occur.  The effect is very short-lived but long enough for the water to be dried out of the image without the drops coalescing, beading or otherwise disrupting the image.

The blanket is heated from below to around 150C.  In addition, after the print station external air blowers pass hot air over the image to further aid drying.  The polymers in the ink remain tacky at this temperature, aiding the transfer process.

There are many claimed advantages to this process.  First of all, it uses aqueous-based inks which are environmentally friendly, and compared to UV-curable inks use relatively low-cost materials.

The freezing of the ink drops on the blanket enables very thin image layers to be produced, typically 500-800 nm.  This means the image takes up the surface features of the substrate, so that an image on a glossy surface will be glossy, and matt on a matt surface, matching the non-image areas.

When the drops are flattened on impact with the blanket and then frozen they retain the shape formed on impact, hopefully circular.  There is no irregular bleed that is sometimes encountered with direct ink jet printing.  In addition inter-colour bleed is also reduced.

The transferred thin film is less likely to split compared to thicker films, reducing the chances of ink being retained on the blanket and thus requiring cleaning.

Optically clear nano-particle ink from Hewlett Packard

US 2013/0222496 A1
Optically clear fluid composition
Hewlett-Packard Development Company, L.P.

Uniform gloss on paper substrates requires a transparent ink to raise the gloss level of the chosen media to the same level of the coloured components of the ink.  If such an ink is to be used then there is a good argument for printing it over the full printed image.  This helps to improve mechanical properties, such as scratch and rub resistance, and potentially introduces other benefits such as UV absorption, therefore increasing the stability of the printed image.

This patent discusses a clear ink jet ink designed to be printed over the full image to equalise gloss, improve mechanical properties and absorb UV, without increasing the overall haze of the printed image.

The novelty of the patent is to use metal oxide nano-particles with a particle size below 140 nm.  This small particle size is necessary for transparency of these materials.  The patent suggests that it is also beneficial if the metal oxide average particle size is similar to that of the colorant pigment particles, in order to stack efficiently and therefore minimise coating roughness and haze.  Such nano-particulate materials also show benefits with respect to stability and printhead performance.

Additionally, these metal oxide dispersions should be compatible with the printed inks in order to prevent aggregation of these nano-particles which might increase haze.  Generally, such metal oxide dispersions are electrostatically stabilised and can be formulated to be stable at a pH of 8-10, the range for ink series such as HP 38 and HP 70.  Specifically the pH of the metal oxide dispersion should be within 0.5 pH units of the coloured inks.

The specific metal oxide particles used should be transparent to visible light and have a refractive index greater than 1.65.  Zinc and aluminium oxide work well for this technology, whereas silica (with a refractive index of 1.45) does not, as it apparently gives hazy coatings.

Exemplified coating formulations contain around 2 wt% of a polyurethane binder, 2 wt% (dry solids) of the metal oxide particles, 10 wt% of humectants (1,2-hexanediol, 2-pyrolidone or similar), ~0.5 wt% surfactants and biocides, with the remainder water.  The mechanical properties of such coatings were compared to a silver halide print, giving equal or better scratch resistance and very low haze levels.

Careful manufacturing process for improved latex ink

US 2012/0232215 A1 
Single batch latex ink compositions and methods 
Hewlett Packard Development Company, L.P.

The addition of polymers to a pigmented ink jet ink will generally improve printed image robustness at the expense of a higher ink viscosity which, depending on the printhead technology used, may thus preclude the use of such materials. As frequently discussed here, polymer encapsulated pigments potentially offer the best of both worlds without any compromise in the ink’s viscosity.

However, even with the best encapsulated pigments, as with everything, there is still some compromise and that comes in the form of a more involved manufacturing process, and it is this point that the patent hopes to address, with a relatively simple approach.

The example used (4.6 wt%) of the monomers styrene, hexamethylacrylate and acrylic acid at a ratio of 40:55:5, and added 0.1 wt% of azobisisobutyronitrile (AIBN) and 2.3 wt% of a pigment powder (BASF D7086 cyan). This paste was then added to water (92.3 wt%) in which had been dissolved sodium dodecyl sulphate (0.7 wt%). This was then mixed at high shear in a microfluidiser at 10,000 psi for 15 minutes to produce the dispersed pigment emulsion.

This emulsion was then charged into a reaction vessel and degassed under argon before heating to 80C for two hours. After this time a portion of 0.1 wt% of potassium persulfate, dissolved in a small amount of water, was added over the course of three hours, with a further 3 hours to ensure full reaction of the monomers. The resultant dispersion was allowed to cool, neutralised with potassium hydroxide and filtered.

The main point exemplified in the text is that the addition of the thermal initiator over time is critical, as without this approach the particle size is generally significantly below 250 nm and unsuitable for ink jet, presumably due to viscosity issues. Thus by starting polymerisation with a relatively low number of initiator species, then the number of polymer particles nucleated will be relatively low and these particles will then grow during the subsequent polymerisation and further nucleation will be minimal. In comparison, if all the initiator is added at the start of the polymerisation, then there will be significantly higher polymer particle nucleation resulting in more, smaller particles in the final ink.

Emulsion-based thermal ink jet inks

WO 2012/105950 A1
Emulsion-based thermal inkjet inks
Hewlett-Packard Development Company

Thermal ink jet technology requires an ink to explosively boil when the relevant firing resistor is activated, and thus print a droplet.  This generally requires inks that use carrier liquids, such as water, that boil at relatively low temperatures.  There are a huge range of applications that require the jetting of fluids that do not boil, either because their boiling points are too high at atmospheric pressure, or they do not have a clearly defined liquid/gas phase transition.

So although piezoelectric printheads are more expensive to produce, they are able to jet a significantly larger range of ink jet ink chemistries than the cheaper, but more limited, thermal printheads.  Some progress has been made in which UV curable formulations have been jetted through thermal printheads, but these have used water soluble UV monomers as they still require water (or some similar organic solvent) to form the bubble.

Using water as a solvent for such aqueous soluble UV curable materials has its own problems, one being that as the concentration of monomers is increased and the proportion of water decreased, the first order phase transition to the gas phase can be smeared out and so the bubble production is impacted.  Ideally more typical hydrophobic monomers would be used in thermal printheads, but as these show very limited water solubility they have seen limited success.

This patent presents an approach to allowing a wide range of high boiling point, hydrophobic solvents, to be used in thermal printheads without resorting to low boiling co-solvents as the drive fluid.  The inventive step is to formulate an ink that is mainly a hydrophobic, high boiling liquid with water dispersed in this continuous phase as a stable water-in-oil emulsion.  It is suggested that other low boiling solvents could be used in this way, but water is the main target.


The figure left demonstrates the theory behind this approach.  A surfactant 12 is added to the high boiling point solvent at an amount above its critical micelle concentration (CMC).  Then, in the absence of water, the surfactant will form micellar structures 14 in which the surfactant polar groups 12a are situated on the inside of a spherical structure.  Small additions of water will hydrate the core 14a of the micelles to give the structure 18, but at this point the water is effectively dissolved in the surfactant and so will not possess bulk water properties.  As the water content is increased, there is a gradual transition to a vesicular structure 20, where a water core 20a is surrounded by a surfactant shell.  In this state the water present possesses its bulk properties and if the vesicle size is small enough it can form a stable dispersion in the continuous oil phase.  Ideally this stable state is a micro-emulsion.

The patent suggests that it has been experimentally shown that 5% water in such a micro-emulsion is enough to jet a hydrocarbon oil-based ink or commercially available UV ink.  In one example 5 wt% of an anionic surfactant sodium bis(2-ethylhexyl) sulfosuccinate (AOT) and 7.7 wt% water were added to Isopar L and mixed to give a micro-emulsion of water-in-oil.  A magenta ink was then formulated by the addition of 2 wt% pigment and this was successfully jetted through a thermal printer, giving a drop weight of 10 pl and a drop speed of 7.5 m/s with a 25V/1.7µs drive pulse.

A similar example took an HP Scitex FB221 light magenta ink and added 8 wt% AOT, 8 wt% water, 8 wt% Isopar L and 15 wt% lauric acid and mixed to give a micro-emulsion.  The lauric acid was included as a viscosity modifier to bring the viscosity below 4 cps.  This formulation was again jetted through a thermal printhead giving a drop weight of around 10 pl and a drop velocity of 7.5 m/s using a drive pulse of 24 V at 3 µs and a precursor pulse of 0.5 µs.

Although these examples seem somewhat preliminary, they seem to prove an interesting point that complex inks, viscosity allowing, could be jetted through thermal printheads.  It will be interesting to see if more specific and complete patents will be published using this chemistry technology to allow the printing of other complex and functional inks through thermal printheads.