Influence of Drop Shape on Printing Performance of CIJ and DOD Technologies
1. Background Information
Domino Printing Sciences was founded in 1978 since then it has fast become a market leader in the field of inkjet printing. To date Domino products are sold in over 120 countries by 75 distributors all over the world, through a network of 17 subsidiaries. Their primary market is coding and marking which involves non-contact printing at high operational speeds. Domino's range of printers and inks are used in many markets including, food and diary, beer and beverage, cosmetic and pharmaceutical, electronic, automotive and household. This means that there must be a large ink range with varying properties and formulations in order to adhere to the variety of substrates used out in the field. From best before dates on eggs where the inks are regulated by heath and safety legislation to batch numbers on plastic cables which need to lightfast so they do not fade over time. Domino products need to be reliable and capable of operating in sometimes dirty, dusty or wet conditions, as any down time on a production line can be very costly to a company.
Companies often have their own set of requirements that Domino products need to fulfil, for example inks may need to be resistant to cleaning process or be safe to be used in a factory where food is being packaged. Domino and other companies in the marking and coding market are finding it increasingly difficult to achieve these more complex requirements asked for by companies for example, faster printing speeds, more economical and eco-friendly. By gaining an increased understanding in the ink jet process it is hoped that this will become easier.
This project is aimed to try to fully understand how printer design and fluid properties influence drop shape and printing performance in CIJ and DOD technologies.
The Basics of Continuous and Drop in Demand Ink Jet Printing.
Continuous ink jet printing involves a single jet of ink which is deliberately made unstable and broken into droplets in a controlled manner.  At the point of jet break up a charge is applied to each droplet as it passes through a charge electrode. The charged drops then pass between two deflector plates. Ink which is not deflected returns to the gutter for recycling to the ink reservoir and the charged deflected droplets form the printed image on the substrate. This process goes on continuously even when the printer is not required to print.
The first practical continuous ink jet printer, based on Rayleigh's break-up, was released in 1952 by Elmqvist of the Siemens- Elma company. However, this was not used as a printer but as medical voltage recorder where the deflecting of the drops was driven through analogue voltages from a sensor.  It was not until the 1960's when Sweet demonstrated that by applying a pressure wave pattern to an orifice, the ink stream could be broken into droplets of uniform size and spacing. When drop break-up was controlled, the drops could be charged selectively and reliably as they emerged out of the continuous ink stream. The charged drops were deflected, when passing through an electric field, to form an image on the substrate. The uncharged drops were captured and re-circulated in the system.  This discovery led to the first commercial CIJ printing product being released by Videojet in 1968.
DOD printing involves a printhead with open nozzles and channels lined with piezoelectric crystals. When an electrical pulse is applied to the piezo crystal on the wall of the channels they change in order for a droplet of ink to be released.  The other types of DOD printing, thermal ink jet and bubble jet, involve the piezo crystal being replaced with a heater element.
The first commercial DOD device, released in the 1960s, was the electrostatic pull device.  This involved conductive ink being held in a nozzle by negative pressure, when a high voltage pulse was applied to an electrode located outside the nozzle a charged droplet of ink was pulled out. By applying the appropriate deflection field the droplet can be located on the substrate. However, in the 1970s the electrostatic pull technology was abandoned as it proved to be unreliable and produced poor print technology. The four different DOD printer technologies that are used today are said to be based on patents that were released in the 1970s.  The first squeeze mode printer was released in 1972, the first bend mode printer was released in 1976 and 1984 the push and shear mode printers were released. Another technology that was found to be useful in DOD printing was sudden steam printing which was discovered in the 1960s. This involves boiling aqueous ink at certain time instances and drops being generated. However, this idea was abandoned until the 1970s when Canon and Hewlett Packard picked it up and released there own printers in 1979 and 1981 respectively. These were the first thermal ink jet technologies.
Domino primarily specialised in Continuous Ink Jet Printing but over recent years it has increased its product portfolio to include ranges for Drop on Demand printing, Thermal Transfer Printing and Laser technology. The Continuous Ink Jet range includes the A-series single jet printers and the binary Bitjet printer with 256 jets, which is used for high speed printing. The Drop on Demand ranges are the K and C-series, these are used for larger characters and outer case coding. The Thermal Transfer printers, V-Series, use ribbons of ink which are melted onto the desired substrate using a headed printhead.
DOD printing is said to be a simpler technology than CIJ as it does not require break off synchronisation, charging electrodes, deflection electrodes, re-circulation systems, high pressure ink supplies and complex electronic circuitry.  However, the techniques needed to make the printhead for DOD i.e. designing the very fine nozzles are more demanding. 
There are many differences between CIJ and DOD printing which are listed in the table below. 
Viscosity of ink used (cps)
Surface Tension ( N/m)
Throw distance (mm)
Print Speed (m/min)
Up to 1000 m/min.
<40m/min (high resolution)
>40 m/min (low resolution)
Dry time (seconds)
The aim of this literature review is to understand the process of continuous ink jet and drop-on demand printing and the fundamentals that affect them. Projects are being carried out at Domino to do this in order to make formulating new ink easier. This is in conjunction with research being performed through an Ink Jet Consortium which been formed between some universities and industrial partners to try and understand these technologies further. The Universities involved are Cambridge, Manchester, Leeds, Durham and Aberystwyth. The industrial partners are CDT, Domino, FFEI, Ince, Linx, Sericol, Sunjet and Xaar.
3.0 The Formation of Liquid Drops
Scientists have investigated liquid drop formation for many years. The first mention of drop formation in literature was by Mariotte in 1686. He was looking at a stream of water flowing from a hole in a container and notes that it decayed into drops. He assumed that it was gravity or other external forces that were responsible for the process.  It was Savart in 1833, who, after using a light to illuminate the jet discovered tiny undulations growing on the jet which grew large enough to break the it. He was the first to recognise that the break-up of liquid jets was governed by laws which are independent of the circumstance under which the jet is produced.  His research showed that break-up always occurs independently of the direction of gravity, type of fluid, jet velocity and jet radius and so must be a fundamental property of the fluid motion and the instability of the jet originates from tiny perturbations applied to the jet at the nozzle opening.  However, he made no reference to the role that surface tension plays in drop break-up; this was recognised in 1849 by Plateau. He showed that disturbances of long wavelengths that lead to a reduction of surface area are favoured by surface tension. In 1879 Lord Rayleigh added flow dynamics to the description of the break-up process. He calculated that a balance of inertia and surface tension sets the time scale for the motion of a jet.  However, Raleigh's linear stability calculation of a fluid cylinder only describes the initial growth of instabilities as they initiate near the nozzle. It doesn't describe break up further down the stream and the processes that lead to satellite production.
Drops are formed in CIJ printing by disturbing a cylinder of ink; these disturbances will grow until the jet breaks. It was Plateau in 1849 who showed how an infinite cylinder of fluid with radius r, disturbances with wavelength, λ > 2Πr will reduce the surface energy and the disturbances will grow.  It was following Rayleigh's discovery, realising the effects of surface tension on the growth of disturbance that most rapid growth happens when λ ≈ 9r.   It is from this work that the driving frequency in CIJ printing is determined as frequency = λv, where v is the jet velocity and λ ≈ 9.  This formula is still used to this day.
3.1 Drop Formation in CIJ Printing.
Continuous ink jet printing involves a single jet of ink that is recycled around the printhead. The ink is supplied into the gunbody under pressure and through a nozzle on the bottom with either a diameter of 60 or 75µm. Break up is caused by “Capillary Instability” which is when a jet will naturally break up into droplets. A drive rod inside the gunbody of the printhead vibrates at 64 kHz causing the jet to break up into 64, 000 drops a second.  As these drops break up they pass through a charge electrode and are given a predetermined charge. The stream of charged droplets then pass between two high voltage plates and are deflected out of the jet stream by being attracted to the positive plate by the negative charge they carry. These drops form the printed image with some drops being charged more than others so an image can be built up. The way an image is built up depends on the raster selected. A raster is a code that defines the number of drops per given stroke. Any drops that are not charged stay in the drop stream and are returned to the gutter for recycling. Sweet said that the best ink for a CIJ system is one that combines (in addition to good marking qualities) high surface tension, low viscosity and high conductivity. 
Schematic of a CIJ printhead
3.2 Drop Formation in DOD
Drop on demand printing involves a print head with an array of nozzles that eject ink only when required.  There are four different technologies used in DOD printing. These all involve piezoelectric units that convert an electrical driving voltage into a mechanical deformation of and ink chamber. This generates the pressure needed to form a single drop. Four different types of printheads are in circulation; squeeze, push, bend and shear with each describing the method in which the piezoelectric causes the ink chamber to deform. The technology most commonly used today is the shear method where the strong shear deformation component in piezoelectric materials is used to deform an ink channel and a drop to be released.
Classification of piezo inkjet printhead technologies by the deformation mode used to generate the drops. 
3.3 Parameters of Ink Jet Printing.
As mentioned before the dominant forces that control the behaviour of the ink jets and drops are surface tension and viscosity. Scientists define these conditions by a series of dimensionless groups developed to compare and analyse the jetting and break up phenomena in Newtonian fluids.  
These dimensionless numbers are:
The Reynolds number, Re:
This describes the ratio between the inertial and viscous forces in a fluid with dynamic viscosity η, density ρ, at a velocity V and a characteristic length D, the drop diameter.
The Weber Number, We:
This describes the ratio of kinetic energy and surface energy, as σ is the surface tension.
The Ohnesorge Number, Oh:
This describes the ratio of viscosity, μ, to the surface forces with L being the characteristic length of the jet. As the jet velocity has been removed.
As ink jet printing requirements become more demanding non-Newtonian fluids are starting to be investigated for their suitability as inks.  The dimensionless group that is used to include the effects of viscoelasticity in non-Newtonian fluids is the Weissenberg number, Wi, where λ is the characteristic relaxation time of the fluid. 
4. Fluid properties
An area of interest recently in ink jet printing is the effect of concentration and molecular weight of polymers used in the ink on the printing performance. This has to be treated with care as addition of polymer to an ink has a strong impact on the nature of the drop generation and ejection process.  Experiments have been carried out by many scientists to see how making changes to the ink alters its performance.   
Evans et al believe that through careful polymer selection the droplet forming properties of an ink can be varied independently of viscosity and so optimised for a particular application. This is why there has been a lot of interest in
G.D. Martin et al say that the chemistry of the ink can disrupt drop formation. For example, the incorporation of long chain polymers tends to inhibit drop break-off, leading to very thin and hence high resistance ligaments. 
4.1 Effects of Polymer Concentration.
D. Xu et al and I.W. Hamley studied the influence of added polymer on drop formation and filament break up as functions of concentration, molecular weight and architecture.  When considering the effects on concentration the dilute regime up to the coil overlap concentration, c*, is investigated. I.W. Hamley defines the coil overlap concentration as the point where individual polymer chains in solution are just in contact. 
From these studies four different regimes have been observed in inkjet drop generation behaviour as a combined function of concentration and molecular weight.  The first regime is said to take place at very low concentration and/or molecular weight and a long ligament is formed that breaks up along its axis to form several satellite drops. This is believed to be a highly chaotic regime and leads to poor print quality. The second regime happens when the concentration or molecular weight is increased further and only a few satellites occur. Regime three is understood to be the optimum regime as with the increased concentration or molecular weight only a single drop is formed. At high concentration or molecular weight the polymer solution becomes visco-elastic and the droplet does not detach, this is regime 4. Although D. Xu agrees with these definitions he states that the concentration and molecular weight range over which the four regimes are identified is highly dependant upon the nature of the polymer, its molecular weight, architecture and the thermodynamic quality of the solvent. 
Evans et al. performed experiments to investigate the effects of polymer concentration and molecular weight in Drop on Demand printing. It was seen that the velocity of the drops decreased with increasing viscosity, the effect was seen more for the higher viscosity inks compared to the low viscosity inks. This was rationalised by considering the polymer concentrations in these inks, there is more polymer present in the high viscosity inks than in the low one, so changes due to differences in polymer will be more evident. The decrease in velocity is said to be as a result of higher molecular weight polymers having more entanglement and coiling so it is harder for the droplet to break away from the bulk liquid and will leave with less kinetic energy.  D. Xu however says that when at the coil overlap concentration the ligament velocity and length were found to only depend on the voltage applied to the piezoelectric. 
4.2 Effects Caused By Changing the Molecular Weight of the polymer
For a higher molecular weight polymer there will be more entanglement/coiling, so it will be harder for the droplet to break away from the bulk liquid, and it will leave with less kinetic energy. 
4.3 Viscoelastic Effects of the Polymer
Suzuki et al say that the jet forming process is associated with viscoelastic properties of jet ink and so rhelogical properties of the inks are investigated. This was confirmed by V. Tirtaatmadja et al who said that understanding the effects of non-Newtonian fluid properties including shear thining and elasticity on drop formation and break-up is of current interest.
Both Suzuki and V. Tirtaatmadja carried out experiments to investigate the influence of the viscoelasticity of polymers on break up and drop formation.   These observations indicated that there were elastic forces present due to the polymer chains acting as springs and/or entanglement effects. It was seen that in low viscosity fluids the elastic forces are irrelevant at the break up point in the jet stream as the dynamics are said to be controlled by an inertial-capillary balance. As a result of this contraction occurred more in inks with higher molecular weight polymer. 
4.4 Effects Due to Temperature Change
Sweet investigated the effects of temperature and noted that it is a critical factor in ink jet printing as it affects the important ink properties of viscosity and surface tension and thus the critical process of drop formation. As viscosity also effects the jet velocity and hence the drop size and deflection sensitivity any changes in temperature which will be detrimental to the viscosity have to be avoided. 
5. Nozzle Influence
It has been said that for a printer to work properly and reliably it must be fitted with a nozzle that is able to produce drops at stable and reliable frequencies, be insensitive to small variations in the fluid parameters and operating conditions and give the shortest break up length possible at the lowest excited voltage. 
B. Lopez et al have said that the influence of nozzle geometry on continuous inkjet break up has been overlooked and seldom considered in the case of capillary instablility- the dominant mechanism of CIJ jet break up. It was first investigated by McCarthy and Molloy who attempted to form a qualitative correlation between the shape of the nozzle and the jet shape produced.  Their research showed that the major requirement of a nozzle is the efficient conversion of potential energy to kinetic energy and is best achieved by a sudden but smooth contraction of the flow area from the supply line to the desired nozzle diameter. There has also been some evidence to show that rounding and polishing the internal nozzle surfaces is favourable in order to achieve a maximum break up performance. 
McCarthy and Molloy continue and say that the nozzle aspect ratio has a significant effect on the initial jet velocity profile and the subsequent jet surface shape and the variability of the ratio is a way of controlling the jet surface roughness. However B. Lopez et al and G. Luxford do not agree and say that the nozzle entry shape has much less effect on the jet velocity and drop break off than may be expected.  
They found that the differences attributed to nozzle geometry might be more related to the flexibility and vibration of the nozzle plate.
6. Satellite Formation.
G. D. Martin et al say that when a stream of drops is created in CIJ printing it is common to find that as well as the principal drop smaller satellite drops can be formed from the ligament as it parts. In DOD printing as the droplet emerges from the nozzle it is observed that it is followed by a ligament or tail which is still connected to the ink in the nozzle. This ligament is said to then part with some being returned to the nozzle and the rest joining the drop or possible breaking up into smaller satellite drops. 
6.1 When satellites occur.
Eggers says that the understanding of satellite drops and their possible control is the driving force behind the research on drop formation.  They were first said to have been observed by Plateau in 1849 in his experiments using an unstable cylinder of liquid in another.  It was Rayleigh, in 1882, who saw the small drops appearing with the main drops and called them “spherules”. 
There have been many different explanations and theories given for satellites production however, is said that these theoires do not give a sufficient explanation.  Rayleigh first produced a linear analysis for drop formation but it has since been questioned by Pimbley et al 1977 as it does not predict the formulation of satellites at all. They have since produced a non linear theory of drop formation which includes the formation of satellites.  This theory uses a temporal instability and neglects the radial inertia and viscous effects. However it is said that this nonlinear theories always predicts that the satellites will separate from both ends of the drop stream at the same time, which is not seen in experimental conditions.
6.2 Types of satellites
Mutoh defines satellites in different modes; rear-merging, infinite, forward-merging and non satellites.  They are used to describe the order of drop formation at both ends of the jet stream and the condition of satellite merging with the main droplet. Rear- merging satellites are said to be formed when the downstream end of the ligament detaches before the upstream end forming a drop with the satellite still attached to the jet stream. This satellite then detaches but is always caught up by the main stream. Infinite satellites are a result of both ends of the stream separating simultaneously and the satellite never catching up with the main stream. Forward- merging satellites are formed when the upstream end of the jet stream separates first forming a drop with a tail. This tail then separates at the downstream end but catches up further down in the jet stream. Non-satellites occur when the forward merging satellite formed when the upstream ligament breaks first and does not separate at the downstream end. Varying the jetting conditions, simulation conditions and physical properties of the liquid used is said to make each of these modes. 
Mutoh goes on to define three regions in a drop formation period which he expresses in phases. The first region is the satellite interaction region which is the phase interval between the separations at both ends of the jet stream. The forbidden region of the main droplet is the phase interval where the resistance to break up is extremely high immediately before the separation at the upstream end of the jet in forward merging satellites. The final region is the forbidden region of the satellite droplet where the resistance becomes extremely high immediately before the separation at the upstream end of the jet in rear- merging satellites.
During the satellite interaction time, when the main drop and satellite are still attached they are said to be drawn towards each other by surface tension forces. These forces are believed to be what causes the drop and satellite to merge further down the jet stream. The distance from drop break up to the merging point of the main and satellite droplet is called the catch-up distance.  The catch-up distance is said to be influenced by the width of the satellite interaction region in which momentum transfer between the main and satellite droplets takes place.
6.3 Problems satellites cause
In continuous ink jet printing the production of satellites has to be controlled to avoid disruption of the printing process. 
Satellite droplets that enter the deflection plates can cause problems for both the main drop and the printer itself. The movements of a charged satellite droplet within the deflection plates are influenced by air drag and electrostatic force and can interfere with the main drop.  The reason for satellites affecting the print quality is explained by G. D. Martin et al; when a satellite enters the electrostatic deflection field it will have a higher charge to mass ratio and will be deflected more strongly which is said to cause build up on the plates. This build up often leads to printer failure. It is postulated by Mutoh that if satellites are to be formed the preferred type is forward-merging as long as they catch up with the main drop before entering the defection plates. 
7. Charging effect on drop shape.
The charging process in continuous ink jet printing, as described by G. L. Fillmore can influence heavily the printing performance.  Before a drop of ink breaks off from the jet stream, it is either selectively charged and used to form part of the character or it returns to the gutter for recirculation. The electrostatic charge is applied to the drop as it passes through the charge electrode. The level of a voltage pulse applied to the charge electrode determines the amount of charge required, and that level is approximately proportional to the desired drop position relative to the character baseline on the printing surface.  When the voltage is applied to the electrode, a negative charge is induced on the drop and is retained after break off until the drop hits the substrate. G. D. Martin et al define the conductive jet, the forming drops and the charging electrode as an R-C circuit in which the resistance and the capacitance both charge with time.  The resistance increases as the drop ligament diameter diminishes and is said to become infinite at the point of break off.
7.1 Charging Interactions
Although the drops are formed individually and they continue in the jet stream alone until they reach the substrate they can be affects in various ways. The charge on a forming drop is said to be not only influenced by the charge electrode potential but also by the charges on previously formed drops.  Because of the capacitance coupling between drops, the charge on the drop about to separate is influenced by the charge on the just previously released drop. Charges on drops further downstream also influence the charge on the separating drop to an extent that decreases rapidly with distance from the charging point. 
The decreased velocity of a stream of drops due to air drag eventually causes the drops to merge unless acted upon by other forces which can cause drop placement errors. Aerodynamic effects on a stream of drops are further complicated by the fact that only the first drop in the stream is travelling through still air.  The remaining drops are said to experience air turbulence and reduced drag due to the passage of their predecessors. Ways to compensate for aerodynamic interaction and for mutual repulsion among drops in flight have been developed by adjusting the charge electrode voltage. 
The forces due to surface charge on a drop oppose those of surface tension and tend to cause the drop to disintegrate. 
In a stream of drops travelling in single file, deceleration due to air resistance is reduced because each drop follows in the wake of the preceding drop. 
7.2 Drop Deflection
Once the drops have left the charge electrode they move through an electric field and are deflected to a height at the paper plane that is approximately proportional to their charge level. [5, 7] The deflection, d, can be approximated by the equation below:
where q is the charge on the drop, m is the mass, E is the electric field, v is the velocity of the drop in the vertical direction, a is the length of the deflection plate and b is the distance from the bottom of the charge electrode to the substrate.  However, this equation is shown to be an under-estimation of the true defection as it does not take into account aerodynamic retardation which slows the drop or field fringing which provides forces at the top and bottom of the deflector plates. 
1. Jens Eggers, “A Brief History of Drop Formation”.
2. Herman Wijshoff, Oc´e Technologies B.V., Venlo, “Structure- and fluid-dynamics in piezo inkjet printheads”.
3.W.T Pimperly and H.C. Lee, “ Satellite Droplet Formation in a Liquid Jet”. IBM Journal of res Development, 1977, 21, 21,
5. Graham D Martin, Stephen D Hoath and Ian M Hutchings, “Inkjet printing - the physics of manipulating jets and drops”.
6. Masayuki Mutoh, “A study of the drop formation of continuous liquid jet by an electrical method”.
7. G. L. Fillmore, W. L. Buehner and D. L. West, “ Drop Charging and Deflection in an Electrostatic Ink Jet Printer”.
8. D. Xu et al, “ Inkjet printing of polymer solutions and the role of chain entanglement”.
9. I.W Hamley, Introduction to Soft Matter, Wiley, New Your, 2000
10. J.J. Cooper-White, J.E. Fagan, V. Tirtaatmadja, D.R. Lester, D.V. Boger, J. Non-Newtonian Fluid Mech, 2002, 106, 29
11. Christopher M. Evans, James E. Fox, Kevin P.Hall and Paul D. Goodwin, “Optimisation of Ink Jet Droplet Formation Through Polymer Selection” .
12. Lord Rayleigh, On the Capillary Phenomena of Jets, Proceedings of the Royal Society of London, 1879, 29, 71-97
13. B. Lopez, A. Soucemarianadin and P. Attane, Break-up of Continuous Liquid Jets: Effect of Nozzle geometry, Imaging Science Technology, 43, 145.
14. Geoffrey Luxford, Momentum Fluctuations in Break-up of an Ink-Jet. Proc. IS&T, NIP, 1999,15, 26-30,
15. R. G Sweet, “High Frequency Oscillography with Electrostatically Deflected Ink Jets,” Stanford Electronics Laboratories Technical Report No. 1722-1, Stanford University, CA (March 1964)
16. The Printing Ink Manual
17. B.L Scheller and D.W. Bousfield, AIChE Journal, 1995, 44, 1357-1367
18. G. H. McKindley, Society of Rheology, Rheology Bulletin, 2005, 74(2), 6-9
19. J. Eggers, Rev. Mod. Phys., 1997, 69 (3), 865
20. M.J McCarthy and N.A. Molloy, Chem. Eng. Journal, 1974, 7, 1.
21. V. Tirtaatmadja, G.H. McKinley and J.J. Cooper-White
22. Suzuki et al.