Tribochemistry and Superlubricity Induced by Hydrogen Ions

Oct 19, 2012 - It is found that the running-in process and low friction state are closely ... Friction Properties of Soft Organosilica Networks for So...
0 downloads 0 Views 3MB Size
Download PDF
Article pubs.acs.org/Langmuir

Tribochemistry and Superlubricity Induced by Hydrogen Ions Jinjin Li, Chenhui Zhang,* Liang Sun, Xinchun Lu, and Jianbin Luo State Key Laboratory of Tribology, Tsinghua University, Beijing, 100084, China ABSTRACT: Friction behavior of aqueous solution at macroscale is quite different from that at nanoscale. At macroscale, tribochemistry usually occurs between lubricant and friction surfaces in the running-in process due to a high contact pressure, and most such processes can lead to friction reduction. In the present work, we reported that the hydrogen ions in aqueous solution played an important role in tribochemistry in running-in process (friction reducing process), which could result in the friction coefficient reducing from 0.4 to 0.04 between Si3N4 and glass surfaces at macroscale. It is found that the running-in process and low friction state are closely dependent on the concentration of hydrogen ions in the contact region between the two friction surfaces. The lubrication mechanism is attributed to tribochemical reaction occurring between hydrogen ions and surfaces in the running-in process, which forms an electrical double layer and hydration layer to lower friction force. Finally, the running-in process of H3PO4 (pH = 1.5) was investigated, which could realize superlubricity with an ultralow friction coefficient of about 0.004. after a running-in period.14−16 A possible mechanism is the tribochemical reaction of silicon nitride or silicon carbide with water, which can not only form a smooth surface for hydrodynamic lubrication, but can also produce colloidal silica for boundary lubrication.17 In our previous work, an ultralow friction coefficient (below 0.01) between a glass plate and a Si3N4 ball was obtained under the lubrication of phosphoric acid solution (pH = 1.5) after a running-in period of 600 s.18 According to the literature cited above, the aqueous solution can also present excellent lubrication properties in macroscale cases studied with a tribometer. However, a running-in period is always required to achieve the ultralow friction. Therefore, it is necessary to investigate the running-in process, which must be helpful to reveal the effective lubricating mechanism of aqueous solutions. In this work, the friction behavior (especially in the runningin process) of electrolyte aqueous solutions with different pH values between Si3N4 ball and glass plate at a high pressure (a maximum contact pressure of 700 MPa) was investigated. The dependence of friction behavior on the pH value in the running-in period was revealed. The lubricating mechanism of aqueous solutions was discussed based on the tribochemical reaction between hydrogen ions and friction surfaces in the running-in process. Finally, the running-in period of superlubricity of H3PO4 was investigated, which could be controlled by the pH value and the volume of solution simultaneously.

1. INTRODUCTION Investigations on the friction behaviors of an aqueous solution confined between two contacting surfaces are very important to develop a water-based lubricant that would be an effective way to partially solve environmental issues. Therefore, the friction properties of aqueous solution at nanoscale have been extensively studied by atomic force microscopy (AFM)1−8 and surface force apparatus (SFA).9−12 For example, Feiler et al. have investigated the dependence of friction on the net interaction force between surfaces in electrolyte solutions with different pH values using AFM.13 Donose et al. have investigated the friction force between a silica particle and silica wafer in pure water and in electrolyte solutions, finding that the adsorbed layers of smaller and more hydrated cations have a higher lubrication capacity than the layers of larger and less hydrated cations.4 Klein’s group has measured the shearing force of aqueous salt solutions confined between two mica surfaces by SFA.11 The excellent lubrication is attributed to forming hydration layers of adsorbed cations, which can keep the two friction surfaces apart and act as a highly efficient lubricant simultaneously. However, when the aqueous solutions are employed as lubricant at macroscale (for instance, load is greater than 1 N and velocity is greater than 1 mm/s), the aqueous solution will exhibit different properties because the mechanical action cannot be neglected under the high contact pressure. Therefore, the lubricating mechanism in nanoscale is not fit for this case anymore, and the tribochemical reaction induced by friction may become an important mechanism for effective lubrication instead. A typical example is the water lubrication with ceramic materials, such as Si3N4/Si3N4 and SiC/SiC. In all cases, an ultralow friction coefficient (below 0.01) is achieved © 2012 American Chemical Society

Received: June 29, 2012 Revised: October 17, 2012 Published: October 19, 2012 15816

dx.doi.org/10.1021/la303897x | Langmuir 2012, 28, 15816−15823

Langmuir

Article

2. EXPERIMENTAL MATERIALS AND METHODS

case. It is also inferred that tribochemical reaction probably occurs between H2SO4 solution and friction surfaces to reduce friction. To investigate whether the negative acid radical ions or positive hydrogen ions contribute to friction reduction, the other acid solutions such as phosphoric acid, hydrochloric acid, lactic acid, oxalic acid, citric acid and sulfamic acid were all used as lubricants between a glass plate and a Si3N4 ball, as shown in Table 1. It is found that the six kinds of acid solutions can all

All the chemical agents used in the test were commercial products with a high purity (>99.5%) and were dissolved or diluted by deionized water (resistivity >18 MΩ·cm) before tests. In this work, sulfuric acid (H2SO4) solutions with different pH values (0, 0.5, 1, 1.5, 2, 2.5, 3.0) were employed to study the effect of hydrogen ions in friction reduction. Other acid solutions, including the solutions of phosphoric acid, hydrochloric acid, lactic acid, oxalic acid, citric acid, and sulfamic acid, with a pH value of 1.0 were also used for comparing. In addition, experiments with sodium chloride (NaCl) solution with a concentration of 5% (pH = 7.0) and sodium hydroxide (NaOH) solution with a pH value of 14 were carried out as a contrast. The friction tests were performed using a Universal MicroTribotester (UMT-2, CETR) in ball-on-disk mode. A Si3N4 ceramic ball with a diameter of 4 mm and a glass plate with a surface roughness of 5 nm were used as the friction pair. Both the ball and glass plate were cleaned in acetone and ethanol for 15 min each in an ultrasonic bath, followed by washing in deionized water and drying by compressed air before test. Before applying load, a few droplets (about 200 μL in total) of the aqueous solution were supplied between the ball and the glass plate. Then the ball was loaded on the glass plate with a force of 3 N, which could generate a maximum contact pressure of 700 MPa. The rotation speed of the glass plate was 180 rpm to provide a linear sliding speed of 0.056 m/s at a track radius of 3 mm. The accuracy of the friction force sensor is 2.5 mN, which makes it accurate enough to measure a friction coefficient value in the 0.001 order with a load of 3 N. All tests were performed at ambient temperature of 25 °C, and the relative humidity was in the range of 25−35%.

Table 1. Friction Coefficient after the Running-In Period under the Lubrication of Six Different Kinds of Acid Solutions name of acid

formula

pH

Trin

friction

phosphoric acid hydrochloric acid lactic acid oxalic acid citric acid sulfamic acid

H3PO4 HCl C3H6O3 H2C2O4 C6H8O7 H3NO3S

1.0 1.0 1.0 1.0 1.0 1.0

150 180 100 190 100 160

0.04 0.03 0.04 0.05 0.06 0.05

lead to friction coefficients reducing to less than 0.08 after a running-in period, although they have different acid radical ions. This result indicates that the hydrogen ions in the lubricant play an important role in friction reduction. To further investigate the relationship between the hydrogen ions and friction reduction, the lubricating behavior of H2SO4 solutions with different pH values (from 0 to 3 with an interval of 0.5) was investigated. As shown in Figure 2a, all the friction coefficients reduce to less than 0.1 after the running-in process

3. RESULTS AND DISCUSSION 3.1. Friction Results. Figure 1 shows the friction coefficient between the glass plate and the Si3N4 ball as a function of time

Figure 1. Friction coefficient with time under the lubrication of H2SO4 solution (pH = 0), NaCl solution (pH = 7), and NaOH solution (pH = 14).

under the solutions of H2SO4, NaOH, and NaCl. It is found that the H2SO4 solution presents the best lubricating property. At the beginning of the test, the friction coefficient reduces from 0.4 to 0.1 rapidly, and then it tends toward stability. Here, we defined the time that elapses from the test beginning to the moment that the friction tends to stability as the running-in period (Trin), which is 50 s for H2SO4 at pH 0. However, such friction reduction can not be found in tests lubricated with NaCl solution and NaOH solution. The friction coefficient remains larger than 0.4 till the end of test. Because the viscosity of H2SO4 solution is close to that of water, it is hard to form the hydrodynamic films under high load and low sliding velocity according to Hamrock−Dowson theory.19 It may be expected that boundary lubrication is the dominant mechanism in this

Figure 2. (a) Friction coefficient with time under the lubrication of H2SO4 solution with different pH values (from 0 to 3 with an interval of 0.5). (b) The relationship between Trin and the pH value of H2SO4 solution. 15817

dx.doi.org/10.1021/la303897x | Langmuir 2012, 28, 15816−15823

Langmuir

Article

in the cases of pH value no more than 2. However, when the pH value of solution is larger than 2, the friction coefficient remains high (above 0.5) until the end of the friction test. In addition, the running-in period is dependent on the pH value of the solution; that is, the higher the pH value, the longer the running-in period, as shown in Figure 2b. These results suggest that the concentration of hydrogen ions in the solution is closely linked to the running-in period, and it has no effect on friction reduction if the concentration of hydrogen ions is very low. After the friction test under the lubrication of H2SO4 solution with pH = 1.0 (the friction coefficient reduced to about 0.05), the friction pairs were unloaded to separate the ball and the glass plate. After that, the friction pairs were loaded to 3N again at the same running track for another test, and the rotation speed and sliding direction remained the same as those in previous tests. It is found that the friction coefficient reduces to a low value of about 0.05 suddenly without a running-in period and then remains stable until the end of the test, as shown in Figure 3. However, as we changed a new running track and

Figure 4. (a) Friction coefficient with time under the lubrication of H2SO4 solution after deionized water was added in the lubricant continuously. (b) Friction coefficient with time after H2SO4 solution (pH = 0) was added in the lubricant.

Interestingly, such a phenomenon is reversible. As the pH value increased to 2.4 and the friction coefficient jumped to about 0.2, H2SO4 solution (pH = 0) was added in the running track to decrease the pH value of the lubricating solution. As the pH value decreases to 0.5, the friction coefficient reduces to the original value (0.05) rapidly in a short period of 15 s, as shown in Figure 4b. Therefore, it is thought that the low friction state is closely related to the concentration of hydrogen ions in the contact region between two friction surfaces. If the concentration of hydrogen ions is high enough, the low friction state can be achieved, whereas the friction would become high if the concentration is too low. 3.2. Friction Surface Analysis. When the running-in process of the lubrication of H2SO4 solution (pH = 1.0) ended, both the ball and glass plate were washed by deionized water for about 15 min, and then their contact regions were measured by a white light interfering profilometer (MICROXAM-3D), as shown in Figure 5a,b. It is seen that the top region of the ball is worn down to flat with a diameter of about 300 μm, and the running track on the glass plate is worn down to concavity with a width of 300 μm and depth of 1 μm. It suggests that the original surface layer is grinded off by friction force during the running-in process. The details of the wear region of ball and the track on the glass plate were also investigated by a scanning electron microscope (SEM, QUANTA 200 FEG), as shown in Figure 5c,d. It is found that there are some nanoparticles with different sizes adsorbed on the track surface, which are mainly abrasive particles from the glass plate according to energy dispersive Xray analysis (EDAX). In addition, some cracks can be observed in the track, as shown on the top right corner of Figure 5d,

Figure 3. Friction coefficient with time after changing the ball, changing the glass plate, and without changing. The lubricant tested is H2SO4 solution with pH = 1.0.

carried out the test with the used ball and glass plate, a runningin process (about 100 s) was required to reobtain the low friction coefficient. Similarly, a running-in process (about 160 s) was also needed as we changed a new ball but ran the test without changing the running track. Comparing with the three results, it suggests that a running-in process is necessary for obtaining the low friction if either or both of the friction surfaces are original. A designed experiment was also carried out to verify the contribution of hydrogen ions to the low friction state. First, the test was run with the lubrication of H2SO4 solution (pH = 1.0) for about 600 s to achieve a low friction coefficient of about 0.05. Second, deionized water was added in the lubricant continuously to reduce the concentration of hydrogen ions in the contact region gradually (the load and rotation speed kept constant all the time). The curve of the friction coefficient with time is shown in Figure 4a. It is observed that the friction coefficient remains constant (about 0.05) until the pH value of the solution rises to 2.4. When the pH value reaches 2.4, the friction coefficient jumps to about 0.25 suddenly, and then fluctuates in the range of 0.2−0.4. When the pH value rises continuously to 3.1, the friction coefficient jumps again to about 0.6, and then keeps a high value. 15818

dx.doi.org/10.1021/la303897x | Langmuir 2012, 28, 15816−15823

Langmuir

Article

Figure 5. (a) The three-dimensional morphology of the top region of the ball. (b) The three-dimensional morphology of the track on the glass substrate. (c) The SEM image of the top region of the ball. (d) The SEM image of the track on the glass substrate.

which indicates that mechanical action has destroyed some regions of the glass surface severely in the running-in process. There are also some abrasive particles deposited on the wear region of the ball surface, but no cracks or pits are observed. To investigate whether the wear on the friction surfaces only occurs in the running-in process, the width of the running track at different moments in the whole friction test (lubricated by H2SO4 solution with pH 1.0) was measured, as shown in Figure 6a. It is found that the width of the track at the beginning of test is calculated to be about 100 μm according to the Hertz contact theory and increases to 300 μm gradually during the running-in process, but it nearly remains constant after the running-in process. This indicates that the wear mainly occurs in the running-in process, leading to the contact region between two friction surfaces becoming larger. After the running-in process, there is almost no wear on the friction surfaces, and the contact region also remains constant due to the low friction. The surface roughness in the track increases from 5 to 11 nm after the running-in process. The average contact pressure reduces to 42 MPa after the running-in process, consequently, due to the increase in the contact area. It is about one-ninth of its original value (at the beginning of test), which becomes a favorable factor for friction reduction. The relationship between the width of the track after the running-in process and the pH value of solution was also investigated, as shown in Figure 6b. It is found that the width of the track after the running-in process is dependent on the pH value of solution. As a result, the contact area increases from 31400 μm2 to 82916 μm2 with increasing the pH value from 0 to 2, which would lead to the average contact pressure reducing from 95 to 36 MPa. This suggests that as the concentration of hydrogen ions in the solution is lower, a larger contact region is required to reduce the contact pressure for realizing low friction after the running-in process. For comparison, the width of the wear tracks on the glass plates was measured after running the tests lubricated by deionized water and NaCl solution for 150 s.

Figure 6. (a) The width of the running track with time in the whole friction test. (b) The relationship between the width of the track after the running-in process and the pH value of solution.

Both the results are about 370 μm in width (corresponding to a contact area of 107466 μm2), which is much bigger than that lubricated by H2SO4 solutions for the same test period. Therefore, it is thought that the wear in the running-in process 15819

dx.doi.org/10.1021/la303897x | Langmuir 2012, 28, 15816−15823

Langmuir

Article

effective surface potential, and e is the electronic charge. The value of ψ0 can be calculated by the Grahame equation as follows:

mainly comes from the mechanical action, and the H2SO4 solutions do not play a promoting role for wear. 3.3. Discussion on the Mechanism of Tribochemistry. According to the results above, it is concluded that there is probably an interaction between hydrogen ions and friction surfaces in the contact region during the running-in process, which is a key factor for friction reduction. To determine the relationship between them, time-of-flight secondary ion mass spectroscopy (ToF-SIMS) analysis was performed on the glass surface. Before ToF-SIMS, a friction test of the glass surface lasted for about 300 s under the lubrication of deuterated phosphoric acid (D3PO4) with a pH of 1.0, and then it was washed by deionized water for 10 s. The intensities of several peaks (including H, D, OH, and OD) in the wear track and outside of the wear track are shown in Table 2. It is found that

σ=

8εε0kBT sinh(eψ0/2kBT ) ×

c(H +)

(3)

where σ is the surface charge density, εε0 is the dielectric constant of solution, and c(H+) is the molar concentration of hydrogen ions in the solution. Here, it is assumed that all sites of SiOH have adsorbed hydrogen ions, and then the mean surface charge density is 0.8 Cm−2 (corresponding to one electronic charge per 0.2 nm2). The relationship between the repulsive pressure and the distance of two surfaces is calculated according to eqs 2 and 3, as shown in Figure 7.

Table 2. ToF-SIMS Analyses in the Wear Track and outside of the Wear Track after a Friction Test with Deuterated Phosphoric Acid in the wear track outside of the wear scar

H

D

OH

OD

29413 11159

539 148

864453 34006

6087 344

there are many deuterium elements on the glass surface even after washing the glass, and the numbers of deuterium elements in the wear track is more than that outside of the wear track. The result indicates that the hydrogen ions in the acid are adsorbed on the friction surface firmly during the friction test. According to the ToF-SIMS analyses, it is considered that when the SiO2 surfaces are brought into contact with an acid solution, the hydrogen ions can be adsorbed on their surfaces by protonation reaction, which is represented as SiOH + H+ ⇒ SiOH+2

Figure 7. The relationship between the repulsive pressure and the distance of two surfaces for different pH value (0.5, 1, 1.5).

It is found that the repulsive pressure increases fast with reducing the distance of two contact surfaces. Additionally, the repulsive pressure increases faster as the pH value is lower. For the case of pH = 1, the repulsive pressure can reach 0.7−5.1 MPa as the distance of the two contact surfaces is in the range of 1−3 nm. Although the calculated repulsive pressure is less than the contact pressure in the tests, it can bear a part of the load and lower the direct contact pressure consequently. When the distance of two contact surfaces are very close (below 1 nm), eq 2 is not fit for this case anymore due to the stern layer effect. The stern layer would prevent the two surfaces from contacting directly because of the strong adsorption force between counterions and surface charge. In this case, the counterions can form a hydration layer to produce a steric hydration force to bear the load.25 Besides these, the direct contact between the surface asperities also possibly exists in the contact region to bear the rest of the load. A possible lubrication model is illustrated in Figure 8. The contact region between two friction surfaces contains three kinds of contacts: direct contact, stern layer contact, and liquid contact. The shearing strength of direct contact is high, but the shearing strength of the stern layer contact (hydration layer) and liquid contact (double electrical layer) is very low, which is the main contribution to the friction force reduction. Therefore, it is concluded that the low friction state obtained after the runningin process is attributed to the electrical double layer effect and the hydration effect on the positive charged surfaces, which can provide a repulsive force as well as a low shearing strength.26 At the beginning of tests, reaction 1 occurs, but the numbers of effective active sites of chemical reaction are limited, and the density of positively charged surface sites is low consequently. Moreover, the original contact pressure is also higher than 300

(1)

where the symbol “” implies that the surface atoms are attached to the underlying bulk solid.20,21 Both the glass surface and Si3N4 surface have SiOH bonds, and surface protonation reaction 1 would occur in the acid solution, which can form many positively charged sites on the friction surfaces. However, when the concentration of hydrogen ions is very low, the ionization of SiOH bonds tends to dominate, which would make the surface become negatively charged. So only when the pH value is lower than pHpzc (a pH value of zero surface charge, for example, ∼2 for silica at 25 °C22), the numbers of positively charged surface sites would be more than the numbers of negatively charged surface sites, leading to the surface electrically positive. In addition, the numbers of positively charged surface sites increases with reducing pH value away from the pHpzc, which would lead to the charging density becoming higher.23 In the electrolyte solution, the surface charge is balanced by the equal dissolved counterions that form the stern layer. The other ions are redistributed and spread away from the surface to form the diffuse electrical double layer. It can produce a repulsive force as the two charged surfaces are approaching. According to the double layer theory,24 the repulsive pressure between two charged surface separated by a distance D is given by the following expression: P = 64kBTρ tanh2(zeψ0/4kBT )e−kD

(2)

where kB is the Bolzmann’s constant, T is the room temperature (298K), k is the Debye screening length of solution, ρ is the number density of hydrogen ions in the solution, ψ0 is the 15820

dx.doi.org/10.1021/la303897x | Langmuir 2012, 28, 15816−15823

Langmuir

Article

3.4. The Running-In Process of Superlubricity with H3PO4 Solution. It should be mentioned that, during the tests, the water in the acid solutions evaporated gradually, which resulted in the increase of pH value of the solution. What would happen after free water in the solution evaporated out completely? As shown in Figure 9, it is found that the friction

Figure 8. The schematic illustration of the lubrication model. The figure in the dashed circle is the illustration of protonation reaction on the friction surfaces. Panels A, B, and C are further illustrations of the three kinds of contacts marked by A, B, and C.

Figure 9. Friction coefficient with time under the lubrication of H2SO4 solution (pH = 1.2), H2C2O4 solution (pH = 1.3), and H3PO4 solution (pH = 1.5).

MPa due to the contact area being small. Therefore, the hydrogen ions have little effect on friction reduction at the beginning. During the running-in process, the original surface layer is grinded off by mechanical action (wear), which could generate many dangling bonds (Si__O−). The dangling bonds (Si__O−) can also react with hydrogen ions in acid solution as follows:27 SiO− + 2H+ ⇒ SiOH+2

coefficient would rise to about 0.2 after free water evaporates out completely for H2SO4, and it would also rise to 0.35 after free water evaporates out completely for oxalic acid. In fact, most kinds of acids have the same phenomenon as these two types of acids, all leading to a high friction once free water has evaporated out completely. However, it is interestingly found that the H3PO4 solution (pH = 1.5) can realize an ultralow friction coefficient of about 0.004 (it is defined as superlubricity when the friction coefficient is below 0.0128) after free water evaporates out completely. This uniqueness of H3PO4 shows its excellent lubricating properties as a lubricant. To further investigate the running-in process of superlubricity of H3PO4, different volumes (10 μL, 20 μL, 30 μL) of H3PO4 solution were supplied on the glass plate before test, and the friction result is shown in Figure 10. It is observed that

(4)

The nascent surface generated by wear has a higher state of surface energy, which would increase the numbers of effective active sites of chemical reaction. So, the numbers of positively charged surface sites would also increase. In addition, the mechanical action can increase the actual contact area between two friction surfaces as the wear scar enlarges. It can not only reduce the contact pressure, but can also increase the numbers of positively charged surface sites in the contact region. All of these factors contribute to lowering friction coefficient in the running-in process. According to the experiment results and discussions above, it is thought that the running-in process is an unstable state between mechanical action and chemical reaction. The mechanical action can produce a larger contact region between two surfaces and lower contact pressure, which can also increase the numbers of chemical reaction sites. It would increase the concentration of charged surface sites and produce a bigger electrical double layer repulsion and hydration repulsion. As a result, the proportion of direct contact region between two surfaces would decrease, which would reduce the mechanical action conversely. These two actions interact with each other, resulting in the friction coefficient reducing gradually until the running-in process ends. After the running-in process, the friction force would be in a stable state, and the contact pressure would reduce to a constant value. If the pH value is lower, the concentration of positively charged surface sites is higher, and the Debye length also becomes shorter,24 both of which would produce a bigger repulsion between the two surfaces, and the mechanism action is less needed. As a result, it needs less time for friction reducing to a stable state, which leads to a shorter running-in period.

Figure 10. Friction coefficient with time under the lubrication of H3PO4 solution with different initial volume (10 μL, 20 μL, 30 μL).

if the initial volume of H3PO4 solution is larger, it needs a longer period to realize superlubricity. For example, when the initial volume is 20 μL, the friction coefficient reduces to 0.05 after about 200 s, and then remains stable until 420 s pass. Finally, the friction coefficient gradually decreases to 0.004 after about 500 s, and keeps this value until the end of test. 15821

dx.doi.org/10.1021/la303897x | Langmuir 2012, 28, 15816−15823

Langmuir According to this result, the running-in process of superlubricity is divided into two stages. The first stage is the friction reducing sharply process, and it is found that the period of this stage is almost the same that for the three initial volumes. In other words, the period of the first stage is independent of the initial volume of lubricant. This process is the common process for all acids, which is a tribochemical process between acids and friction surfaces, as we discussed above. Therefore, the period of the first stage only relies on the pH value of the solution. The second stage is the low friction process, and it is found that the period of this stage is dependent on the initial volume of solution, which becomes longer if the initial volume is larger. Therefore, it is inferred that this stage is associated with the free water evaporating process. At the end of this stage, excess free water in the lubricant has evaporated out completely, and a transparent solid-like film is formed on the sliding track. At that moment, the friction coefficient reduces to 0.004, which enters the superlubricity stage. The ultralow friction is mainly attributed to the hydrogen bond effect that forms a hydrated water layer on the phosphoric acid−water film with a low shearing strength and somewhat dipole−dipole effect, which has been described in detail in our previous work.18 According to these analyses, it is clear that the running-in period of superlubricity of H3PO4 contains two parts. The first part is the tribochemical process between H3PO4 and friction surfaces, and that period depends on pH value of solution. The second part is the free water evaporating process, and that period depends on the volume of solution. Therefore, the running-in period of superlubricity can be controlled by the pH value and the volume of solution. To reduce the running-in period, we can reduce the pH value and the volume simultaneously. Unfortunately, the running-in period cannot be zero, for if the pH value reduces to a definite value (below 0.5) or the volume reduces to a definite value (below 3 μL), the superlubricity cannot be realized.

ACKNOWLEDGMENTS

■

REFERENCES

The work is financially supported by NSFC of China (51075227), the Program for New Century Excellent Talents in University of the Ministry of Education of China, and the Basic Research Program of Shenzhen (0021539012100521066).

(1) Marti, A.; Hahner, G.; Spencer, N. D. Sensitivity of frictional forces to pH on a nanometer scale: A lateral force microscopy study. Langmuir 1995, 11, 4632−4635. (2) Vakarelski, I. U.; Brown, S. C.; Rabinovich, Y. I.; Moudgil, B. M. Lateral force microscopy investigation of surfactant-mediated lubrication from aqueous solution. Langmuir 2004, 20, 1724−1731. (3) Yan, X. P.; Perry, S. S.; Spencer, N. D.; Pasche, S.; De Paul, S. M.; Textor, M.; Lim, M. S. Reduction of friction at oxide interfaces upon polymer adsorption from aqueous solutions. Langmuir 2004, 20, 423− 428. (4) Donose, B. C.; Vakarelski, I. U.; Higashitani, K. Silica surfaces lubrication by hydrated cations adsorption from electrolyte solutions. Langmuir 2005, 21, 1834−1839. (5) Oncins, G.; Garcia-Manyes, S.; Sanz, F. Study of frictional properties of a phospholipid bilayer in a liquid environment with lateral force microscopy as a function of NaCl concentration. Langmuir 2005, 21, 7373−7379. (6) Onoe, H.; Gel, M.; Hoshino, K.; Matsumoto, K.; Shimoyama, I. Direct measurement of the binding force between microfabricated particles and a planar surface in aqueous solution by force-sensing piezoresistive cantilevers. Langmuir 2005, 21, 11251−11261. (7) Polat, M.; Sato, K.; Nagaoka, T.; Watari, K. Effect of pH and hydration on the normal and lateral interaction forces between alumina surfaces. J. Colloid Interface Sci. 2006, 304, 378−387. (8) Yilmaz, H.; Sato, K.; Watari, K. Normal and lateral interaction of α-alumina in different electrolyte solutions. Mater. Sci. Eng. B: Adv. Funct. Solid-State Mater. 2008, 148, 215−220. (9) Klein, J.; Kamiyama, Y.; Yoshizawa, H.; Israelachvili, J. N.; Fredrickson, G. H.; Pincus, P.; Fetters, L. J. Lubrication forces between surfaces bearing polymer brushes. Macromolecules 1993, 26, 5552− 5560. (10) Raviv, U.; Laurat, P.; Klein, J. Fluidity of water confined to subnanometre films. Nature 2001, 413, 51−54. (11) Raviv, U.; Klein, J. Fluidity of bound hydration layers. Science 2002, 297, 1540−1543. (12) Briscoe, W. H.; Titmuss, S.; Tiberg, F.; Thomas, R. K.; McGillivray, D. J.; Klein, J. Boundary lubrication under water. Nature 2006, 444, 191−194. (13) Feiler, A.; Larson, I.; Jenkins, P.; Attard, P. A quantitative study of interaction forces and friction in aqueous colloidal systems. Langmuir 2000, 16, 10269−10277. (14) Tomizawa, H.; Fischer, T. E. Friction and wear of silicon-nitride and silicon-carbide in water - Hydrodynamic lubrication at low sliding speed obtained by tribochemical wear. Tribol. Trans. 1987, 30, 41−46. (15) Chen, M.; Kato, K.; Adachi, K. Friction and wear of self-mated SiC and Si3N4 sliding in water. Wear 2001, 250, 246−255. (16) Zhou, F.; Adachi, K.; Kato, K. Friction and wear property of aCNx coatings sliding against ceramic and steel balls in water. Diam. Relat. Mat. 2005, 14, 1711−1720. (17) Xu, J. G.; Kato, K. Formation of tribochemical layer of ceramics sliding in water and its role for low friction. Wear 2000, 245, 61−75. (18) Li, J. J.; Zhang, C. H.; Luo, J. B. Superlubricity behavior with phosphoric acid-water network induced by rubbing. Langmuir 2011, 27, 9413−9417. (19) Dowson, D.; Higginson, G. R. Elasto-Hydrodynamic Lubrication, 2nd ed.; Pergamon Press: Oxford, U.K., 1977. (20) Sahai, N. Is silica really an anomalous oxide? Surface acidity and aqueous hydrolysis revisited. Environ. Sci. Technol. 2002, 36, 445−452.

4. CONCLUSION In summary, we have studied the friction behavior of acid solution between Si3N4 and glass friction pairs at ambient conditions, showing that the hydrogen ions in aqueous solution can reduce friction by tribochemical reaction. The hydrogen ions can be absorbed on friction surfaces firmly by protonation reaction, which can produce positively charged surfaces. Additionally, the pH value can influence the concentration of charged surface sites, which would result in a difference of running-in period. The low friction state is mainly attributed to the electrical double layer effect and the hydration effect on the positive charged surface that is formed by the adsorbed hydrogen ions. Finally, the running-in process of superlubricity realized by H3PO4 was investigated, which is divided into two parts and can be controlled by the pH value and the initial volume of solution. This result gives a method for reducing the running-in period of superlubricity, which is useful for improving the phosphoric acid solution to be lubricated in actual working conditions in the future.

■

■

Article

AUTHOR INFORMATION

Corresponding Author

*Telephone: +86-10-62773129. E-mail: chzhang@tsinghua. edu.cn. Notes

The authors declare no competing financial interest. 15822

dx.doi.org/10.1021/la303897x | Langmuir 2012, 28, 15816−15823

Langmuir

Article

(21) Sverjensky, D. A.; Sahai, N. Theoretical prediction of single-site surface-protonation equilibrium constants for oxides and silicates in water. Geochim. Cosmochim. Acta 1996, 60, 3773−3797. (22) Parks, G. A.; Bruyn, P. L. D. Zero point of charge of oxides. J. Phys. Chem. 1962, 66, 967−973. (23) Dekeizer, A.; Fokkink, L. G. J.; Lyklema, J. Thermodynamics of proton charge formation on oxides - Microcalorimetry. Colloids Surf. 1990, 49, 149−163. (24) Israelachvili, J. N. Intermolecular and Surface Forces, 2nd ed.; Academic Press: New York, 1991. (25) Leikin, S.; Parsegian, V. A.; Rau, D. C.; Rand, R. P. Hydration forces. Annu. Rev. Phys. Chem. 1993, 44, 369−395. (26) Chen, M.; Kato, K. J.; Adachi, K. The comparisons of sliding speed and normal load effect on friction coefficients of self-mated Si3N4 and SiC under water lubrication. Tribol. Int. 2002, 35, 129−135. (27) Sjoberg, S. Silica in aqueous environments. J. Non-Cryst. Solids 1996, 196, 51−57. (28) Matta, C.; Joly-Pottuz, L.; Bouchet, M. I. D.; Martin, J. M. Superlubricity and tribochemistry of polyhydric alcohols. Phys. Rev. B 2008, 78, 085436−1−085436−8.

15823

dx.doi.org/10.1021/la303897x | Langmuir 2012, 28, 15816−15823