UPTEC K 18024
Examensarbete 30 hp Juni 2018
Boric acid as a lubricating
additive in fuels and in hydraulic oils
Simon Ström
Teknisk- naturvetenskaplig fakultet UTH-enheten
Besöksadress:
Ångströmlaboratoriet Lägerhyddsvägen 1 Hus 4, Plan 0
Postadress:
Box 536 751 21 Uppsala
Telefon:
018 – 471 30 03
Telefax:
018 – 471 30 00
Hemsida:
http://www.teknat.uu.se/student
Abstract
Boric acid as a lubricating additive in fuels and in hydraulic oils
Simon Ström
Boric acid based fuel and oil additives were investigated in this study, with the aims to gain a deeper understanding of how the boric acid fuel additive behaves, to investigate the effect of low rates of fuel additive addition and tribofilm longevity, and to investigate how boric acid behaves as a hydraulic oil additive. Fuel additive experiments were performed in a reciprocating sliding rig with a cylinder on flat contact geometry with fuel additive sprayed on the contact repeatedly, whereas the hydraulic oil experiments were performed in a reciprocating sliding rig with a ball on flat contact with the oil and additive present from the start. Analysis was performed using vertical scanning interferometry (VSI), scanning electron microscopy (SEM) and energy dispersive x-ray spectroscopy (EDS).
The tribofilms created by the fuel additive provided excellent friction reduction capabilities, even with low or no rate of
replenishment. As more additive was sprayed, wear resistance seemed to increase as the surface became increasingly covered. Film coverage need to be less than 20% of the surface in order to gain full
friction reducing effects.
The hydraulic oil additive had little effect on friction or wear resistance under the used parameters and no tribofilm was found.
Examinator: Peter Broqvist Ämnesgranskare: Urban Wiklund Handledare: Staffan Jacobsson
Populärvetenskaplig sammanfattning:
Tribologi, från grekiskans tribo, gnida, är vetenskapen om ytors kontakt och den friktion, nötning och smörjning som uppstår när ytor rör sig mot varandra. Redan när egypterna byggde pyramiderna användes kunskap om tribologi när de använde vatten för att smörja de slädar som de flyttade stenblock med. Senare var den berömde Leonardo da Vinci den första vetenskapsman som studerade friktion, men termen tribologi började inte användas förrän 1966, då forskaren Peter Jost fastslog att friktion och dålig smörjning kostade samhället enorma summor varje år.
Sedan dess har forskning inom detta interdisciplinära fält bidragit till stora effektiviseringar inom en mängd områden. De här effektiviseringarna leder inte bara till stora besparingar av pengar, utan också till minskade utsläpp från fossila bränslen när mindre energi krävs för att driva exempelvis en bil.
Ett område som kan effektiviseras ännu mer är en minskning av friktionsförluster mellan kolvringar och cylinder i förbränningsmotorer. Kontakten här leder till en betydande del av de totala friktionsförlusterna i en motor. Ett annat område är friktionsförluster i hydraulsystem, där friktion leder till värme som kan skada smörjoljorna och i förlängningen också hydraulsys- temet självt.
Vanligen tillsätter man additiv i smörjoljor och i viss utsträckning även i bränsle, för att få förbättrade tribologiska egenskaper. Det kan vara allt från additiv som är antioxidanter och därigenom stabiliserar oljan vid höga temperaturer, till additiv som ska hålla oljan ren. I vissa fall bildar additiven så kallade tribofilmer, skyddande eller friktionssänkande lager på ytan.
Vanliga additiv innehåller zink (Zn), svavel (S) eller fosfor (P), som kan ha problematiska effekter på människor, miljö eller till och med i katalysatorsystemen på bilar.
Ett potentiellt additiv för förbättrade friktionsegenskaper är borsyra. Borsyra är ett fast smör- jmedel med lagerstruktur, som är starkt inom lagren, men svagt och lättskjuvat mellan lagren.
Det gör att borsyra kan ge en låg friktion när rörelsen är parallell med lagerstrukturen, då lagren lätt kan glida mot varandra. Borsyra är dessutom ett vanligt ämne i naturen, därmed billigt, och det är dessutom inte heller giftigt. Tanken är att borsyra kan användas som ett effektivt, billigt och miljövänligt bränsleadditiv för minskad friktion i förbränningsmotorer, och dessutom som ett additiv i hydraulolja för minskad friktion även där. Fältförsök med borsyra som en bränsletillsats för både bensin- och dieseldrivna bilar har gett goda resultat, men inga försök har gjorts med borsyra i hydraulmotorer.
För att undersöka borsyrans smörjande egenskaper kan man använda sig av förenklade mod- eller av de system som borsyran är tänkt att agera i. En enkel, reciprok fram-och-återgående rörelse kan användas för att undersöka friktionen mellan två ytor under många testcykler. En jämförelse av friktionsförloppen för smörjmedel med och utan borsyran ger en enkel bild av borsyrans inverkan, men den är inte fullständig. Kontaktytornas utseende kan undersökas med vitljusinterferometri (VSI) eller med svepelektronmikroskop (SEM), medan energidispersiv rönt- genspektroskopi (EDS) i kan användas för att undersöka ytornas och de eventuella tribofilmer- nas sammansättning och en mer fullständig förståelse av borsyrans effekt som friktionssänkande medel.
Målen med studien var att jämföra bränsleadditivet Triboron, innehållande borsyra, med en trietylboratlösning för att fastställa att Triboron bygger upp en faktisk borsyrafilm. Ett an- nat mål var att undersöka livslängden och underhållsbehovet av de tunna filmer som skapas
när Triboron används. Det gjordes genom att genomföra långa tester där mängden tillförd Triboron varierades. Ett tredje mål var att undersöka de friktionssänkande egenskaperna och nötningsskyddet när borsyra används som en tillsats i hydrauloljor.
Studiens resultat tyder på att de filmer som uppstår när Triboron används är uppbyggda av borsyra och inte något annat borbaserat material. Det visade sig också att filmerna hade god livslängd, även med mycket lågt filmunderhåll, och att en yttäckningsgrad på mindre än 20%
är nödvändig för att uppnå mycket goda friktionssänkande egenskaper. När yttäckningsgraden stiger minskar nötningen, men friktionen påverkas mycket lite. Slutligen påvisades det att under de testparametrar som använts så gav borsyra ingen effekt på vare sig friktion eller nötningsmotstånd i hydrauloljor.
List of symbols and abbreviations B(OH)3 Boric acid
B2O3 Boric oxide EtOH Ethanol
2Bu 2-butoxy ethanol TEB Triethyl borate PAO8 Poly-alpha-olefin 8 RME Rape methyl ester
EDS Energy dispersive x-ray spectroscopy SEM Scanning electron microscopy
VSI Vertical scanning interferometry
ICE Internal combustion engine
Contents
1 Introduction 1
1.1 Background . . . . 2
1.2 Aim of the study . . . . 2
2 Theory 2 2.1 Friction . . . . 2
2.2 Viscosity . . . . 3
2.3 Lubrication . . . . 3
2.3.1 Lubrication regimes . . . . 4
2.3.2 Liquid lubricants . . . . 5
2.3.3 Solid lubricants . . . . 5
2.4 Internal combustion engines . . . . 5
2.5 Hydraulic motors . . . . 7
3 Boron in tribology 7 3.1 Boric acid . . . . 8
3.2 Boric acid as a lubricant . . . . 9
4 Experimental 10 4.1 Comparison of Triboron fuel formula and Triethyl borate . . . . 11
4.2 Effect of reduced spraying . . . . 12
4.3 Hydraulic oil tests . . . . 12
4.4 Analysis . . . . 13
4.4.1 SEM . . . . 13
4.4.2 EDS . . . . 13
4.4.3 VSI . . . . 14
5 Results 14
5.1 Comparison of Triboron fuel formula and Triethyl borate . . . . 14
5.1.1 Friction tests . . . . 14
5.1.2 SEM and EDS analysis . . . . 15
5.2 Effect of reduced spraying . . . . 19
5.2.1 Friction behavior . . . . 19
5.2.2 Method 1, SEM and EDS analysis . . . . 22
5.2.3 Method 2, SEM and EDS analysis . . . . 25
5.2.4 Method 3, SEM and EDS analysis . . . . 28
5.3 Hydraulic oil tests . . . . 31
6 Discussion 36 6.1 Comparison between Triboron and triethyl borate . . . . 36
6.2 Effects of reduced spraying . . . . 36
6.3 Hydraulic oil tests . . . . 39
7 Conclusions 40
8 Further studies 40
References 42
1 Introduction
The word tribology comes from the Greek words tribos and logos, meaning “rubbing” and
“study of”, respectively. It is the study of friction and wear between surfaces in contact, and the lubrication of these surfaces in order to reduce both friction and wear. It is an interdisciplinary field, with contributions from chemistry, physics, materials science and engineering.
In cars, losses to friction have been estimated to 28% of the fuel energy [1], see Figure 1. Of these losses, roughly 35% are due to friction in the engine. Several components contribute to the friction losses in the engine, such as bearings, the valve train, and the piston assembly.
Reducing friction in the piston assembly could reduce the fuel consumption considerably. When a reduction in friction losses is achieved, less fuel need to be burned, which mean that the cooling and exhaust losses decrease accordingly. Thus, a reduction in friction losses gives a total fuel consumption reduction roughly triple the friction loss reduction. The friction losses in cars equaled about 208,000 million liters of gasoline and diesel in 2009, which is approximately 340 liters per car per year. If the total friction losses could be decreased by 18% the savings would be more than 117,000 million liters of gas and diesel per year, which in turn would decrease CO2 emissions by 290 million tonnes yearly [1].
Figure 1: A breakdown of average fuel consumption of a passenger car. A reduction of friction losses in the engine and transmission would lead to a better fuel economy [1].
Hydraulic motors use large amounts of oil, that to some extent ends up in the environment as pollutants due to unintended spill. The Swedish forest industry alone is responsible for 40,000 m3 of oil and lubricant pollutions every year [2], as the hydraulic machines operate at high pressures, high flow rates and in a harsh environment. Therefore, a switch to greener, biodegradeable hydraulic fluids is interesting. However, high temperatures in contacts due to friction in the hydraulic motors may lead to the degradation of these fluids. A reduction of friction could increase the service life of the fluid, and consequently make greener fluids more attractive.
1.1 Background
The company Triboron International AB manufactures a fuel additive containing boric acid, B(OH)3, that has been shown to decrease fuel consumption in passenger cars by 5.5% on average [3], and by 8-12% in diesel generators [4]. Boric acid is a polar molecule, and can therefore be difficult to use in oils due to agglomeration of the boric acid particles [5], but the additive from Triboron solves this problem, and forms a stable dispersion in an ethanol solution, where the boric acid is present in the form of triethyl borate. When used as a fuel additive in internal combustion engines, the boric acid is believed to be active in the piston assembly, delivered there together with the fuel. This would mean that boric acid is continously added to the surfaces in contact, the piston head and cylinder lining, and is able to continously build up a friction reducing tribofilm.
The study of this boric acid fuel additive, called Triboron fuel formula, has been underway at Uppsala University for some years, but there is still much that is unknown. Earlier studies have resulted in a suitable method for testing the effect of the fuel formula. This method will be used in this thesis.
Since there are positive results when using the boric acid additive in gasoline and diesel, there is interest to see whether the additive has similar friction reducing effects in hydraulic fluids.
In this application boric acid is mixed into the fluid from the beginning, and will not be further added to the fluid. One important difference between lubricating oils and hydraulic fluids is that the hydraulic fluid’s main purpose is to transfer power, not to lubricate the contact.
1.2 Aim of the study
The aims of this thesis are:
• To gain a deeper understanding of how the boric acid in the fuel additive behaves compared to how triethyl borate behaves,
• to investigate the effect of reducing the addition of fuel additive on the friction behavior,
• to investigate how long lasting the boric acid tribofilms are after the addition of fuel additive has decreased or stopped,
• and to investigate how boric acid behaves as an additive in hydraulic fluids and to deter- mine if a stable tribofilm is formed.
2 Theory
2.1 Friction
The resistance to motion that occur when two surfaces in contact move relative to one another is called friction, and is a central concept in tribology. When no lubricant is present the friction is called dry friction, which follows the two laws proposed by Amonton, called Amonton’s first and second laws:
1. The force of friction is directly proportional to the applied load.
2. The force of friction is independent of the apparent area of contact.
In addition to these laws, Coulomb’s Law of Friction states that the kinetic friction is indepen- dent of the sliding velocity. These laws lead to Equation 1, where µ is the friction coefficient, Ff is the friction force and FN is the normal force.
µ = Ff FN
(1)
The static friction coefficient, µs, is generally higher than the kinetic friction coefficient, µk. This means that the friction for a resting object is higher just before the object starts moving than the friction when the object is already moving. In the tribological experiments performed for this thesis, the friction coefficient refers to the kinetic friction coefficient.
2.2 Viscosity
Substances that can flow display a resistance to flowing called viscosity. In a liquid, it is caused by physical entanglement of the molecules, or by intermolecular bonding. Disentanglement is caused by Brownian motion, which increases with temperature, and the breaking of bonds which also increases with temperature. Hence, an increasing temperature will lead to a lower viscosity of a liquid. The viscosity can be seen as an internal friction of a flowing substance; the greater the viscosity, the greater the inner friction. On the other hand, a liquid with a larger viscosity can carry a larger load, as a larger force is needed to deform the liquid.
Introducing a shear stress to a viscous liquid will lead to deformation of the liquid. For New- tonian liquids, the shear stress τ is proportional to the shear rate ˙γ via the viscosity η, see Equation 2.
τ = η ˙γ (2)
Friction between two parallel planes separated by a fluid is given by the shear stress of the fluid and the area A of the planes, see Equation 3 [6]. Oils used for lubricating purposes are generally Newtonian liquids.
Ff = τ A = η ˙γA (3)
2.3 Lubrication
The actual contact between two surfaces is concentrated to the surface asperities, which carry most of the load. Since these asperities are small, the contact pressure tend to be large even at smaller loads, which can lead to high friction or wear. The primary purposes of lubrication are to reduce friction and reduce wear, but several additional secondary benefits, such as reduced
temperatures or reduced corrosion, are also important [6]. These effects are achieved by fully or partially separating the surfaces with a lubricant, such as an oil film or a grease.
2.3.1 Lubrication regimes
Lubrication is classified by three different regimes: boundary lubrication, where the surfaces are only separated by adsorbed molecules on the surfaces, full film lubrication, where the surfaces are fully separated by a lubricating film, and finally mixed lubrication, which is a mixture of boundary and full film lubrication. These regimes are represented in Figure 2. A shift from boundary, to mixed, to full film lubrication occurs when the film pressure builds up sufficiently to shift the contact pressure from the asperities to the film.
Figure 2: The three lubrication regimes. a) boundary lubrication, b) mixed lubrication and c) full film lubrication [7].
In boundary lubrication, a) in Figure 2, the load is still carried by the asperities. The surfaces are, however, separated by a thin layer of lubricant. The lubricant can be a solid, liquid or gas, and is of a molecular thickness, adsorbed on the surface. The direct contact between the surfaces is reduced or completely hindered by the film and the friction is determined by the sliding resistance of the film [6]. In lubricated contacts the highest friction levels are usually seen in boundary lubricated contacts.
Mixed lubrication, b) in Figure 2, is similar to boundary lubrication in some aspects. A protec- tive film is present between the asperities here as well, and they still carry a large part of the load, but an increasingly large component of the load is shifted to the lubricant. The friction is determined by both the viscous shear of the lubricant and the sliding resistance of the protective film.
Full film lubrication, c) in Figure 2, occurs when the surfaces are fully separated by the lubricant.
The separation results in very good protection from wear, and with a suitable lubricant the friction will be low, determined by the force needed to shear the lubricant [6]. There are several types of full film lubrication, such as hydrostatic or hydrodynamic lubrication. Hydrostatic lubrication requires the use of an external pump to pressurize the oil so that the surfaces are not in contact. Hydrodynamic lubrication is created by the movement of a surface, dragging lubricant into a decreasingly small gap, thus pressurizing the oil. In a hydrodynamicly lubricated system the surfaces are in contact while the surfaces are stationary, only boundary lubricated.
Risk of wear is therefore highest at start and stop of the application.
As the surfaces are fully separated during full film lubrication, wear is minimized and friction is low. Full film lubrication is thus a desirable type of lubrication. In practical applications however, mixed and boundary lubrication is becoming more and more dominant due to the decreasing size of machine components, leading to smaller separation tolerances and higher contact pressures [8].
2.3.2 Liquid lubricants
The most common type of liquid lubricants are oils. Oils can be divided into three subgroups:
mineral oils, synthetic oils and vegetable oils [6]. Mineral oils and synthetic oils are both produced from crude oil. Distilling the crude oil allow for the production of mineral oils, while synthetic oils are made by breaking down crude oil components and synthesizing a desired oil.
Vegetable oils are, as the name implies, produced from vegetable or animal products. They are triglycerides, composed of fatty acids [6], and have two properties that affect their usefulness:
they are often unsaturated, and they are polar. Being unsaturated results in a chemically active region, where double bonds can react with other molecules in the air or the oil. Acidic products can lead to corrosion, and products from oxidation can form harmful deposits in the application.
The polar property of vegetable oils means that the molecule can adsorb to a surface, making vegetable oils well suited as boundary lubricants.
2.3.3 Solid lubricants
Another group of lubricants are the solid lubricants. Among these lubricants are ones with lamellar structure, such as graphite. These materials are made up of atomic layers, interacting with each other through weak van der Waal’s bonds. This allow the layers to be easily sheared, resulting in low friction when forces are applied parallel to the planes [9]. The drawback of lamellar solid lubricants is that if they are forced out of the contact and not renewed the lubri- cation is lost. There is an absolute life span of the lubricant - when it is worn down no further lubrication is possible [6]. Examples of lamellar solid lubricants are graphite, molybdenum sulfide or boric acid.
2.4 Internal combustion engines
There are more than one billion cars in the world today [10], the vast majority of them using an internal combustion engine (ICE). Other vehicles, such as ships, or stationary generators often use ICE as well. The purpuse of the ICE is to transform the energy stored in the fuel into useful mechanical work, such as in cars or generators. However, a large part of the energy in the fuel is lost as friction or thermal losses [1]. In a passenger car approximately 21.5% of the fuel energy is used to move the car [1]. Four stroke engines are typically used in cars. The basic principle of such an engine is shown in Figure 3.
The contact between the piston and cylinder wall is further described in Figure 4. The piston and cylinder wall are in contact via the piston rings: loosely fitting rings whose main purpose is
to seal the combustion compartment so that the gas pressure works on the piston fully. Since this is a sliding contact, effective lubrication is needed to reduce friction and wear. This lubrication is chiefly achieved through splashing of oil upon the cylinder walls by the crankshaft.
Figure 3: A four-stroke engine in action. Intake of air and fuel is shown in 1 as the piston moves down. The mixture is compressed, 2, with the upward motion of the piston. The fuel is combusted in 3, drinving the piston downwards a second time. The final upward motion of the piston exhausts the gases, 4. These four piston motions comprise the four strokes of one combustion cycle, hence a four-stroke engine [11].
Figure 4: Piston and cylinder wall contact. The inset to the right show a piston ring in contact with the wall, lubricated by an oil film [11].
2.5 Hydraulic motors
A radial piston hydraulic motor works by converting hydraulic pressure into torque and rotation [12, 13]. The motor consists of a high speed pump, usually driven by an electric motor, feeding hydraulic fluid to cylinders, Figure 5. Fluid at high pressure is fed to the cylinder in the working stroke, whereas low pressure is used in the return stroke. The cylinders house the piston assembly, which is the connection between the stationary housing and the rotating shaft.
Rolling and sliding under high forces take place here, generating heat due to friction which may degrade the hydraulic oil [14].
Figure 5: Example of a radial piston hydraulic motor design. Left: cross section of motor from the side, showing the shaft and two cylinders. Right: The outer cylinder wall has a cam profile, and the inner houses the pistons. Encircled is one of the piston assemblies. [14]
The primary purpose of the hydraulic fluid, often an oil, is not to lubricate the engine, but rather to transfer power. The entirety of the motor interior is submerged in hydraulic fluid so that all moving parts are lubricated during operation. The oil in the motor is contained in a closed system; no oil is added nor (ideally) consumed during operation.
3 Boron in tribology
Boron is a metalloid group III element with symbol B and atomic number 5, a light element at the top of the periodic table, with considerable hardness, 9.5 Mohs. Elemental boron however, is difficult to prepare and is poorly studied and barely used. It has the electron configuration 1s22s22p1, that is, three valence electrons that can be involved in covalent bonds [8]. Boron readily forms oxides, sulfides, nitrides and halides, often adopting planar structures. These three-valent compunds are electrophiles, and in terms of Lewis acid-base theory are Lewis acids as they readily accepts electron pairs from Lewis bases. Due to the intrinsic electron deficiency of boron, its reactivity is mainly determined by its Lewis acidity. The empty p-orbital can be easily attacked by nucleophiles, for example water, as is the case in boric acid.
Boron’s ability to combine with other tribologically interesting elements, such as nitrogen or sulfur, allow it to form compounds that can be used as liquid lubricants, solid lubricants or as lubricant additives [8].
3.1 Boric acid
Two crystalline forms of boric acid exists: orthoboric acid (B2O3 · 3 H2O, H3BO3 or B(OH)3) and metaboric acid (B2O3 · H2O or HBO2), of which orthoboric acid is the most common [15, 16]. Orthoboric acid is a stable, white solid at room temperature [15, 17] and is soluble in water and several organic solvents such as ethanol or 2-butoxy ethanol. The molecule consist of a trivalent boron atom, surrounded by three hydroxyl groups in a planar conformation, see Figure 6. Hydrogen intermolecular bonds between the hydroxyl groups in combination with the planar structure result in solid boric acid having a layered structure [8, 18], as can be seen in Figure 7.
This layered structure is easily sheared and therefore suitable as a solid lubricant [19, 20].
Figure 6: The planar chemical structure of the boric acid molecule.
Figure 7: The layered structure of boric acid. Strong hydrogen bonds are present between hydrogen and oxygen atoms in the layer, whereas only weak van der Waal’s forces are present between layers [11].
A drawback of boric acid as a lubricant is that it loses water at temperatures over 100◦C, forming HBO2 followed by boric oxide (B2O3), Reaction C .1 [8, 21, 22]. Boric oxide lack the layered structure of boric acid and therefore lack the acid’s lubricating properties [23]. Another possible problem with boric acid is that it can be forced out of the contact area as it is a solid lubricant, thus rendering the contact unlubricated [8]. The continuous replenishment of boric acid during operation is important for the lubrication [8, 19].
B(OH)3 −−→ B2O3 + H2O∆ (C .1)
When boric acid is dissolved in ethanol it is present in the form of triethyl borate, Figure 8.
Triethyl borate reacts with water tribochemically to form boric acid [8, 24], see Reaction C .2.
Figure 8: The chemical, non-planar, structure of triethyl borate.
B(OCH2CH3)3 + H2O −−→ B(OH)3 + EtOH (C .2)
3.2 Boric acid as a lubricant
One way of supplying the contact with boric acid is with boric acid particles as an additive in lubricants, done by Lovell et al. [25]. In their study nano- (20 nm), submicro- (600 nm average size) and microparticles (4 µm average) of boric acid was studied as additives in canola oil. In a pin-on-disk test it was found that the nanoparticles gave the best results regarding reduced friction and wear, outperforming the larger particles as well as a MoS2 additive by an order of magnitude, with a coefficient of friction around 0.04. This was explained as the 20 nm particles forming a colloidal dispersion in the oil, allowing the boric acid to easily enter the contact.
It was also found that a mixture of submicro- and microparticles performed better than the oils with either size of particles alone, explained as the smaller particles forming a protective boundary layer while the larger particles act as load support between the surfaces.
Erdemir et al. [26] found that a layer of boric oxide was formed when annealing a B4C substrate at 800◦C. The boric oxide subsequently formed a thin boric acid film when reacting with water in the air. Using a pin-on-disk apparatus, the coefficient of friction for this film was 0.03-0.05.
A base oil containing boric acid was investigated by Baş et al. [15] using a pin-on-disk setup.
They found that using a concentration of 4 wt% boric acid, the coefficient of friction could be lowered by up to 50%, and that the fuel consumption for a 170 kW diesel engine decreased by 3.6%.
A previous study performed at Uppsala University by Larsson et al. [19] investigated different methods of adding the boric acid to a sample. Three methods were investigated: A) repeated spraying of the surface with a boric acid solution, B) a predeposited boric acid film, and C) a mixture of methods A) and B). Using a reciprocating cylinder on flat configuration, the coefficient of friction was investigated. While method C was shown to give the lowest coefficient of friction, lower than 0.020, method A was found to be the most stable method, reaching friction levels of 0.050. The tests were performed at room temperature and at 100◦C, simulating the temperatures at the cylinder wall in a combustion engine. The coefficient of friction was found to be lower in room temperature tests, and the similar behavior of the tests suggested that the room temperature tests can be seen as representative. It was also found that the boric
acid tribofilm had a poor stability, complicating the investigation of the films. The method developed by Larsson et al. is used in this thesis.
Further studies by Larsson et al. [27] attempted to link the friction losses seen in the earlier paper [19] to friction losses of actual field tests [3]. Friction reductions of up to 89% were observed mid-stroke, with lower reductions close to the turning points. It was reasoned that a smaller friction reduction in the piston assembly could lead to a larger fuel saving than expected, due to the previously mentioned reduced losses associated to heat and cooling.
4 Experimental
Two similar test rigs were used for the experiments. One of the test rigs is shown in Figure 9, and was used for the experiments in Sections 4.1 and 4.2. These experiments also used a spray nozzle, shown in Figure 10. The other rig is smaller, and uses a different contact geometry, but all other parts are the same. The smaller rig was used for the experiments in Section 4.3.
The connecting rod and circular plate (7) could be adjusted to set the track length, this rig used 5 cm. The trolley (6) held the sample (5) and moved relative to the stationary cylinder holder (4). The load was variable by changing the weights (2) on the cantilever (3), in this rig a load of 5 N was used. A ventilated hood (not shown) was placed over the rig, and a spray nozzle (8, Figure 10) with a container (9) was aimed at the sample.
Computer software was used to control the spray and the speed of the rig. Data for the friction force was recorded by the load cell (4). Data was collected every 10 cycles and 200 data points were collected per cycle. The data was available as friction curves for individual cycles. The average friction for every cycle was also plotted against time.
Figure 9: Reciprocating test rig: 1) Counterweight, 2) weights, 3) cantilever, 4) cylinder holder and load cell. In the other test rig, a ball holder was used. 5) Sample, 6) trolley, 7) connecting rod and circular plate.
Figure 10: Spray setup: 8) Nozzle, 9) container for solutions.
4.1 Comparison of Triboron fuel formula and Triethyl borate
The experiment was carried out in a test rig with a bearing steel cylinder sliding against a reciprocating cast iron flat surface, with a periodic spraying of Triboron or triethylborate (TEB) solutions.
The cast iron samples were ground with SiC-paper with grit size 120, followed by grit size 1000.
Both grit sizes were used in a cross hatch pattern, resembling the surface pattern of actual cylinder liners. A thin oil film was predeposited on the sample surface by covering the surface with a mixture of polyalphaolefin-8 (PAO8) and hexane and allowing the hexane to evaporate.
The oil mixture was 30 wt% PAO8. A film thickness of 30 µm was used, calculated by Equation 4.
hoil= msample+oil− msample
Asampleρoil (4)
A spray nozzle was used to spray boric acid solution on the sample during the testing, simulating the solution being added to the engine with the fuel. The nozzle was calibrated to spray 30 mg/spray. The solutions used were a Triboron fuel formula, prepared by Triboron, and a TEB solution, prepared as 4 wt% TEB in EtOH.
The test parameters were as follow: a load of 5 N, a frequency of 1 cycle/second and a stroke length of 5 cm so that the total cycle length is 10 cm. A test length of 2000 cycles was used, where the spraying was started after 1000 cycles, to allow the friction to stabilize before spraying.
The solution was sprayed every tenth cycle (every tenth second) for a total of 100 sprays. The friction coefficient for the individual cycles as well as the average friction for each cycle was recorded.
Reference tests with an oil film deposited on the sample and no spraying, as well as with spraying of EtOH were also carried out. The tests are summarized in table 1.
Table 1: Summary of tests performed.
Test Spray
Reference, oil film None Reference, oil film and spray EtOH
Triboron Triboron fuel formula
Triethylborate Triethylborate solution
4.2 Effect of reduced spraying
To test the effect of reduced spraying, the same setup as in Section 4.1 was used. After 1000 cycles without spraying followed by another 1000 cycles of spraying every tenth cycle, the amount of fuel formula sprayed was changed. The fuel formula used was provided by Triboron, containing 4.4 wt% B(OH)3 and 2 wt% rape methyl ester (RME).
A load of 5 N was used, a frequency of 1 cycle/second and stroke length 5 cm and a total of 100,000 cycles. All tests saw a reduced amount of fuel formula sprayed: in Method 1, the fuel formula solution was sprayed every 100 cycles, in Method 2 every 1000 cycles, and in Method 3 a EtOH/PAO8 solution containing no fuel formula was sprayed every 10 cycles. The tests are summarized in Table 2. During Method 1, the nozzle was cleaned every 20,000 cycles by putting it in an ultrasound bath for 5 minutes with hexane followed by 5 minutes with EtOH.
This was done in order to reduce the risk of clogging the nozzle.
Table 2: Summary of tests performed with a reduced level of sprayed Triboron fuel formula.
The amount of fuel formula decreases from Method 1 to 3.
Test 0-1000 cycles 1001-2000 cycles 2001-100,000 cycles Method 1 No spray Every 10 cycles Every 100 cycles Method 2 No spray Every 10 cycles Every 1000 cycles
Method 3 No spray Every 10 cycles Every 10 cycles, no Triboron
4.3 Hydraulic oil tests
The experiment was carried out in a test rig with a bearing steel ball sliding on a reciprocating flat, measuring the friction coefficient. The ball had a diameter of 10 mm. Tests were initially performed on bearing steel samples as it is a similar steel type to the ones used in many hydraulic motors, but as the wear tracks were found to be difficult to investigate further tests were performed on cast iron samples. A fully formulated commercial hydraulic oil, Panolin HLP SYNTH 46, was used. Several different mixtures of this oil with different boric acid solutions were investigated, primarily different in boric acid concentration. Two different solvents for the boric acid were also investigated, ethanol (EtOH) and 2-butoxy ethanol (2Bu). All oils were mixed by Triboron, with different amounts of their fuel formula, containing 8.8 wt% B(OH)3 (EtOH) or 4.4 wt% B(OH)3 (2Bu). The oils investigated are presented in Table 3. The reference oil is the Panolin oil without any boric acid added.
From preliminary tests using the reference oil the following test conditions were determined: a test length of 20,000 cycles, a speed of 4 cycles/second, and a stroke length of 1 cm, resulting in a total sliding distance of 40,000 cm. These parameters were used in all following hydraulic oil
Table 3: Hydraulic oils investigated
Oil Solvent Formula:Oil B(OH)3 (wt%)
Reference - - -
1 EtOH 1:500 0.0176
2 EtOH 1:100 0.088
3 2Bu 1:100 0.044
4 2Bu 1:50 0.088
tests. The normal load was 10 N, equivalent to a mean Hertzian contact pressure of 0.43 GPa.
The oil was predeposited on the sample before starting the test.
4.4 Analysis
The samples were cleaned in hexane before analysis in order to remove the oil. Scanning electron microscopy (SEM) and vertical scanning interferometry (VSI) were used to study the surfaces of the samples, whereas energy-dispersive X-ray spectroscopy (EDS) was used to obtain elemental information of the tribofilms.
4.4.1 SEM
Investigating boric acid films in a SEM can be troublesome due to the non-conducting nature of the films. This leads to charge build up, which in turn leads to a distorted image. To reduce this problem, low acceleration voltages were used, 3 and 5 kV. Tilting the sample reduces the activated volume, further lowering charge build up, and can also be done to improve the imaging.
LEO1550 manufactured by Zeiss, with EDS AZtec was used for the SEM and EDS analysis.
Wear tracks were analyzed and potential tribofilms were searched for.
4.4.2 EDS
Using EDS to detect boron is hard, especially in combination with the low acceleration voltages required. The low acceleration voltage leads to a small activated volume. This means that few photons are emitted and consequently detected. On the other hand, a small activated volume is beneficial when studying films, as the substrate will not be detected. The Kα energy peak of boron (0.1834 keV) is close to that of carbon (0.2774 keV). Furthermore, carbon is an easily detected substance, and peak broadening of the carbon signal can lead to the software mistakenly identifying B where only C is present. In order to confidently determine the boron presence a well distinguished boron peak must be observed.
EDS spectra and mapping was obtained using a 5 kV acceleration voltage, with a 60 µm aper- ture. The EDS analysis was aimed at determining the chemical composition of any suspected tribofilms, mainly done by mapping.
4.4.3 VSI
The instrument used for VSI analysis was the WYKO NT1100 manufactured by Veeco: 0.5X and 1.0X field of view lenses were used together with an 10X objective. The depth and width of the wear tracks were studied with VSI.
5 Results
5.1 Comparison of Triboron fuel formula and Triethyl borate
5.1.1 Friction tests
The results from the friction tests are presented in Figure 11. All tests show a similar behavior for the first 1000 cycles, before spraying starts. The friction first decreases during this time, due to a smoothing of the surfaces in contact, and then levels out. After spraying commenced, the reference test with EtOH spray display no difference compared to the reference without spraying. The friction for the tests including Triboron and triethyl borate display an immediate decrease in friction as a reaction to the addition of their respective solutions to the track.
The curves of the Triboron and triethyl borate tests display spikes, more prominently in the triethyl borate test. These spikes occur every tenth cycle in connection to that the solution is sprayed onto the track. There is a slight friction change during the spray, that is followed by an almost immediate drop back. In the triethyl borate test there is a slight oscillation in the size of these peaks.
Figure 11: Comparison of how different sprayed solutions affect the friction. The two refer- ences have no effect on friction levels, whereas the solutions containing Triboron fuel formula and triethyl borate display decreasing friction levels.
5.1.2 SEM and EDS analysis
The flat sample surfaces were investigated with SEM and EDS to determine the existence of, appearance of, and contents of any formed tribofilms. Figure 12 show SEM images of a sample where Triboron fuel formula was sprayed. The cross hatch pattern from the preparation of the samples can be clearly seen in all four images. In the images, darker areas indicate the presence of a tribofilm. The first image, Figure 12a is taken close to the middle of the track, and is representative of the overall appearance of the track. Larger dark patches can be seen where the ground lines cross, as well as diffuse dark lines parallel to the sliding direction. In the next Figure 12b, the same cross hatch pattern and lines parallel to the sliding direction can be seen.
A larger light patch is also visible where the cast iron substrate shows up. Figure 12c is a close up image of the tribofilm, where some charge up of the film can be seen. The last image, Figure 12d shows the turning point of the track furthest from the spray nozzle. Here, the black area in the lower part of the picture is the build up of a thick film outside the track.
SEM images from a sample sprayed with triethyl borate can be seen in Figure 13. Here, Figure 13a is similar to Figure 12a in that both images are taken close to the center of the track, representing the overall appearance. In this image, a few larger particles are present, and a dark gray color can be seen covering most of the shown area. Following in Figure 13b the large degree of coverage is visible at a greater magnification, whereas Figure 13c show a large patch of cast iron showing through the covering film. Finally, Figure 13d show that the dark lines parallel to the sliding direction are present on this sample as well.
Comparing the images in Figures 12 and 13, the tribofilm coverage is greater on the sample sprayed with triethyl borate. Furthermore, another difference is the grooves from the sample preparation - on the triethyl borate sample only grooves running in one direction are visible.
Figure 14 shows a SEM image along with EDS maps for Fe, C, O and B of a Triboron sprayed sample. The maps show that in the dark areas B and O is present, as well as C, indicating the formation of boric acid along with carbon contaminations. In contrast, the Fe map shows that the cast iron substrate is not visible through the dark areas. The boric acid is present in the grooves created during grinding of the sample, as well as in the dark lines parallel to the sliding direction.
In Figure 15, a sample sprayed with triethyl borate is shown. EDS maps indicated the presence of boric acid in the grooves and in lines parallel to the sliding direction.
(a) (b)
(c) (d)
Figure 12: Flat sample after a test including spraying with Triboron fuel formula. The sliding direction is shown by the arrows. (a) Representative area of the sample, close to the middle of the sample. The cross hatch pattern comes from the grinding during sample preparation. (b) Area with a gray patch without the protective film. (c) Close up of the tribofilm, some charge up of the film can be seen. (d) The turning point of the track, the black area is outside the track. In (a), (b), and (d) darker lines parallel to the sliding direction can be clearly seen. In-lens SEM at 3 kV.
(a) (b)
(c) (d)
Figure 13: SEM images of a sample sprayed with triethyl borate. The sliding direction is shown by the arrows. (a) Area near the center of the sliding track, a large part of the track is covered.
Some larger particles are visible. In (b) a similar area as in (a) is enlarged. (c) Patch of cast iron, surrounded by a dark area. Lastly, (d) shows that lines parallel to the sliding direction are present also on this sample. The cross hatch pattern is dominated by grooves running from the upper left to the lower right. In-lens SEM at 3 kV.
Figure 14: Surface after spraying with Triboron together with corresponding EDS maps for Fe, C, O and B. The arrow indicates the sliding direction. 3 kV.
Figure 15: Surface after spraying with triethyl borate together with corresponding EDS maps for Fe, C, O and B. The arrow indicates the sliding direction. 3 kV.
5.2 Effect of reduced spraying
5.2.1 Friction behavior
Results from the 100,000 cycles friction test are presented in Figure 16. The friction dropped as an immediate reaction when spraying of Triboron fuel formula started at cycle 1000. After the initial 2000 cycles which are common to all tests, the curves all display a similar decrease in friction, reaching approximately 0.045 by 10,000 cycles. Thereafter, Method 1 display somewhat unstable levels of friction, with a few large spikes to high friction levels. However, the friction return to low levels after the unstable regions. Method 2 shows some instability between 20,000 and 50,000 cycles, but return to a stable level after that. During the last 12,000 cycles, smaller peaks can be observed, coinciding with the spraying. Compared to Methods 1 and 2, Method 3 is stable throughout the experiment, although an oscillating pattern can be observed connected to the spraying interval.
Some tests were performed that failed. An observation was made in these tests that in cases where a slightly higher amount of boric acid solution was sprayed the friction rose rapidly after an initial decrease, to the point where the test had to be aborted. Here, large build up of boric acid was observed at the turning points of the track. The build up was highly viscous, more like grease than a liquid.
Figure 16: The effect of reduced levels of sprayed fuel formula. After the 2000 initial cycles (common to all tests) all Methods show a similar decrease in friction, until roughly 10,000 cycles.
The dashed line marks 2000 cycles, before this line the experiments are performed identically.
Figures 17 through 19 show the friction curve during the forward strokes of selected individual cycles. Cycle 1000 is the final cycle before Triboron spraying commence, thus showing the friction when only lubricated with the oil film. Cycle 2000 is the last cycle of the period with Triboron spray every 10 cycles. By cycle 10000 the tests were deemed stable. Lastly, cycle 100000 is the final cycle of the experiments.
Starting with Figure 17, from Method 1, all cycles have in common that the friction is at its highest at time = 0. This correspond to the turning point closest to the spray nozzle. The other turning point, at time = 0.5, also displays higher friction than the middle parts of the track. The decrease in friction mainly occurs due to a decrease in the middle part of the stroke.
However, by cycle 100,000 the friction has increased in the middle of the track compared to cycle 10,000, while the friction at both turning points is lower.
Figure 18 from Method 2 display similar behavior as in Figure 17. Friction is the highest at the turning points, with a lower friction in the middle of the stroke. Here, the lowest levels between cycle 10,000 and 100,000 are similar, however cycle 10,000 has a lower friction in larger parts of the stroke.
Finally, in Figure 19 displaying cycles from Method 3, cycles 1000 and 2000 display very similar levels of friction in the middle of the track. Also here it can be seen that the overall friction is lower in cycle 10,000 rather than in cycle 100,000.
Figure 17: Friction curve for selected forward strokes of individual cycles from Method 1, spray every 100 cycles, after cycle 2000. Friction is at its highest close to the turning points, at 0 s and 0.5 s.
Figure 18: Friction curves of individual cycles from Method 2, spray every 1000 cycles, after cycle 2000.
Figure 19: Friction curves of individual cycles from Method 3, EtOH and PAO8 spray after cycle 2000.
5.2.2 Method 1, SEM and EDS analysis
After 100,000 cycles, samples were investigated with SEM and EDS. Figure 20 shows an image of the track of a sample from Method 1. Several dark lines can be seen parallel to the sliding direction, covering most of the image. In the upper right corner the cross hatch pattern from the sample preparation is visible, this pattern is only vaguely visible elsewhere. This indicates that the sample is covered with a film, that is absent or very thin in this area. Similar to the images seen in Section 5.1, these darker areas appear to be a boric acid film.
Figures 21 and 22 show the two most prominent surface appearances. In Figure 21 the surface is mostly covered by a tribofilm, with nothing of the underlying pattern visible. In contrast, in Figure 22 the cross hatch pattern is clearly visible. The dark areas are concentrated around and in the pattern grooves.
Figure 23 shows a detail of Figure 22 clearly showing that the darker areas consist of a thin film on top of the cast iron substrate. Encircled is one of several flattened areas, acting as sliding plateaus. Such plateaus are locally present on top of the entire film on the sample, both in areas where the substrate is fully covered as well as in partly covered areas, exemplified by the image.
Figure 20: Representative image from Method 1 after 100,000 cycles, sample sprayed with Triboron fuel formula every 100 cycles after a 2000 cycle run in period. The arrow shows the sliding direction. Dark lines can be seen parallel to the sliding direction. The cross hatch pattern is visible in the top right corner, but only vaguely in the rest of the image. In-lens SEM at 5 kV.
Iron can only be seen locally in the wear track, indicating that the tribofilm cover the track to a large degree, see EDS maps in Figure 24. The presence of B and O over almost the entire area suggest that the tribofilm is composed of boric acid. C is also present in large areas.
Figure 21: Method 1: Showing the topography of the wear track from Method 1. The cross hatch pattern is not visible, as the tribofilm fully covers the surface. Grooves along the sliding direction are prevalent. The darker areas likely indicate where the film is thicker. SE2 mode at 5 kV, tilted 40◦.
Figure 22: Method 1: A region where the cross hatch pattern is clearly visible and filled with boric acid. Grooves parallel to the sliding direction are visible on the dark patches. SE2 mode at 5 kV, tilted 40◦.
Figure 23: Method 1: Close up of the central part of Figure 22. Encircled is an example of flattened sliding plateaus present on the film. The cross hatch pattern is clearly visible in the cast iron substrate under the tribofilm. SE2 mode at 5 kV.
Figure 24: Method 1: Fe is only showing in small areas on the EDS maps, whereas B, O and C cover the surface.
5.2.3 Method 2, SEM and EDS analysis
The cross hatch pattern is clearly visible in Figure 25, and the right also shows several dark lines parallel to the sliding direction. The visibility of the underlying pattern indicates that no film has formed, or that the film is very thin. At higher magnification, Figure 26 shows the lighter areas to some degree are covered with a grainy structure. The darker areas appear flatter and more cohesive, indicating that they are sliding plateaus that have become flattened by the cylinder passing over them.
Images taken with a 40◦tilted sample and an SE2-detector can be seen in Figures 27-29. In the first of these, Figure 27, two clearly different areas are seen. Judging by previous images, it is reasonable to believe that the darker areas are covered by the tribofilm to a larger degree than the lighter areas. Both the lighter areas and the darker areas are common on the sample. A close up of an uncovered area is seen in Figure 28. The grooves from the cross hatch pattern have become filled, presumably with boric acid. A large scratch in the cast iron surface is marked by the red arrows. Scratches like this, and smaller were found in the uncovered areas, but not in the covered areas. Figure 29 shows a close up of a darker area. The film has been smoothened as sliding plateaus have formed on the thickest parts. The cast iron substrate can be seen through holes in the film. Where the film is thin the cast iron topography can be seen through the film.
Figure 25: Typical character of the wear track of a Method 2 sample, with Triboron spraying every 1000 cycles. The cross hatch pattern is clearly visible across the entire image, suggesting a limited film coverage. Darker lines parallel to the sliding direction are prominent in the right half of the image. In-lens mode at 5 kV.
EDS mapping shows that the grooves have been filled with B and O, presumably boric acid, see Figure 30. The dark lines from the SEM image are visible in the O and B maps, whereas C is present on almost the entire surface.
Figure 26: Method 2: Close up of Figure 25, showing the presence of a film in the lighter as well as in the darker areas. The darker areas appear flatter and more cohesive. In-lens mode at 5 kV.
Figure 27: Method 2: Two areas can be identified, one darker that is covered to a large degree, and one lighter, not as covered. The cross hatch pattern is visible in both areas, however. Both types are representative of the sample, both highly covered and less covered areas are common.
SE2 mode at 5 kV, 40◦tilt.
Figure 28: Method 2: The cross hatch pattern has been filled. A larger scratch parallel to the sliding direction is marked by the red arrows, smaller scratches are also visible. This type of scratches were otherwise rare on the sample, concentrated to the uncovered areas. SE2 mode, 5 kV, 40◦tilt.
Figure 29: Method 2: Close up of a film-covered area. The film has been smoothened by the sliding cylinder. In some places, the substrate pattern can be seen through the film. SE2 mode, 5 kV, 40◦tilt.
Figure 30: Method 2: The grooves have been filled with a substance containing B and O. Also the dark lines in the right part of the SEM image correspond to slightly increased signals from B and O. C is detected on almost the entire surface. EDS analysis at 5kV.
5.2.4 Method 3, SEM and EDS analysis
Dark lines appear at regular intervals, parallel to the sliding direction, throughout the entire image, see Figure 31. These lines were present over the entire sliding track, displaying the same regularity between the lines. The lines are approximately 20 µm apart, but appear to be spaced slightly wider in the right part of the image. The lines are about 10 µm wide. The cross hatch pattern is clearly visible, with larger dark areas concentrated around intersections of the grooves. By tilting the sample 40◦the lines can be identified as ridges, see Figure 32. The area between the ridges is in some places covered with a dark film. However, Figure 33 shows also the bare substrate in the lighter areas between the ridges.
Figure 34 shows a pair of large scratches commonly found in the track. Furthermore, these particular scratches are about 10 µm apart, coinciding with the width of the ridges in Figure 33. The lower part of the image also contains several smaller scratches.
EDS mapping shows a large presence of B and O in the ridges, indicating that they are composed of boric acid. The cross hatch grooves appear to be filled with the same substance. C is present over the entire surface, slightly more concentrated to the ridges. Fe can be seen clearly between the ridges.
Figure 31: A sample sprayed with an EtOH/oil solution after 1000 cycles of Triboron spray, Method 3. Dark lines at relatively regular intervals can be seen parallel to the sliding direction (shown by the arrow) throughout the image. These lines were present over the whole sliding track. The cross hatch substrate pattern is clearly visible. In-lens mode, 5 kV.
Figure 32: Method 3: The topography of the lines seen in figure 31 can be identified as ridges.
The ridges are at intervals of about 20-30 µm. SE2 mode, 5 kV, 40◦tilt.
Figure 33: Method 3: Closeup of Figure 32, showing the bare substrate between the ridges.
SE2 mode, 5 kV, 40◦tilt.
Figure 34: Method 3: Scratches like these, running parallel to the sliding direction were commonly found in the track. SE2 mode, 5 kV, 40◦tilt.
Figure 35: Method 3: EDS maps of the ridges seen in Figures 31-34 show that they are composed of B, O and C, with the Fe substrate visible between the ridges.
5.3 Hydraulic oil tests
The initial hydraulic oil tests performed on bearing steel substrates are presented in Figure 36. The friction differs between the different oils for the first 12,000 cycles, after which most oils reached the same level of friction, about 0.12. The EtOH 1:500 test was performed using a different calibration of the test rig, which is likely the explanation to why the friction differ from the other oils. It is reasonable to believe that under the same circumstances this friction curve also would stabilize around 0.12, as both the reference oil and the EtOH 1:100 oil do.
The two oils containing the 2-butoxy ethanol solvent both display a slight friction decrease com- pared to the reference, but after about 5,000 cycles the friction increases, passing the reference, before decreasing and stabilizing at the same level as the other oils.
In order to better discern the substrate wear, cast iron substrates were used. Friction curves for these are presented in Figure 37. Here, the initial variation seen during the first 12,000 cycles in Figure 36 is absent, and friction of the three oils quickly stabilize after only a few hundred cycles.
At about 8,500 cycles the oil with 2-butoxy ethanol gave a slight friction increase. However, since the friction is very stable both before and after this increase, it is likely only the effect of a coincidental occurence, such as a wear particle getting stuck on the sliding ball.
Analysis of the cast iron samples shows that there is very little to no presence of boron in the tracks, Figures 38 - 40, regardless of oil used. Compared to outside the tracks, inside all three tracks there is a greater amount of oxygen, and the iron is visible. Figure 38 shows one of the turning points for the track from the reference oil, with the track in the right part of the image.
In Figure 39 several dark spots are seen in the track, shown to be rich in carbon. These spots could be graphite particles in the cast iron.
The topographies of the tested cast iron substrates are displayed in Figures 41 - 43, in a 2D view (a), and 3D view (b) of the same area. In both the 2D and the 3D view it is clear how the tracks have been smoothened to a similar degree in all three tests. The asperities outside the
