journal of the mechanical behavior of biomedical materials 32 (2014) 145–154
Available online at www.sciencedirect.com
www.elsevier.com/locate/jmbbm
Research Paper
Effects of different blasting materials on charge generation and decay on titanium surface after sandblasting Cecilia Yan Guo, Alexander Tin Hong Tang, James Kit Hon Tsoi, Jukka Pekka Matinlinnan Dental Materials Science, Faculty of Dentistry, The University of Hong Kong, 34 Hospital Road, Sai Ying Pun, Hong Kong, PR China
art i cle i nfo
ab st rac t
Article history:
It has been reported that sandblasting titanium with alumina (Al2O3) powder could
Received 11 September 2013
generate a negative electric charge on titanium surface. This has been proven to promote
Received in revised form
osteoblast activities and possibly osseointegration. The purpose of this pilot study was to
21 December 2013
investigate the effects of different blasting materials, in terms of the grit sizes and electro-
Accepted 28 December 2013
negativity, on the generation of a negative charge on the titanium surface. The aim was
Available online 6 January 2014
also to make use of these results to deduct the underlying mechanism of charge generation
Keywords:
by sandblasting.
Titanium Dental implants Sandblasting
Together 60 c.p. 2 titanium plates were machine-cut and polished for sandblasting, and divided into 6 groups with 10 plates in each. Every plate in the study groups was sandblasted with one of the following 6 powder materials: 110 mm Al2O3 grits, 50 mm
Electric charge
Al2O3 grits, 150–300 mm glass beads, 45–75 mm glass beads, 250 mm Al powder and 44 mm Al
Alumina powder
powder. The static voltage on the surface of every titanium plate was measured
Glass beads
immediately after sandblasting. The static voltages of the titanium plates were recorded
Aluminum powder
and processed using statistical analysis. The results suggested that only sandblasting with 45–75 mm glass beads generated a positive charge on titanium, while using all other blasting materials lead to a negative charge. Furthermore, blasting grits of the same powder material but of different sizes might lead to different amount and polarity of the charges. This triboelectric effect is likely to be the main mechanism for charge generation through sandblasting. & 2014 Elsevier Ltd. All rights reserved.
1.
Introduction
Sandblasting is a commonly used surface treatment method for titanium dental implants to improve on their bone-toimplant bonding. Previous studies have mostly focused on n
Corresponding author. Tel.: þ852 285 90380; fax: þ852 254 89464. E-mail address: [email protected] (J.P. Matinlinna).
1751-6161/$ - see front matter & 2014 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.jmbbm.2013.12.026
the mechanical impacts of sandblasting on the titanium surface, such as the created roughness (Papadopoulos et al., 1999), residual stress, subsurface microstructure (Pazos et al., 2010) and eventual contamination (Al Jabbari et al., 2012). Among these effects, modifying the surface roughness of
146
journal of the mechanical behavior of biomedical materials 32 (2014) 145 –154
titanium and its alloys has been considered the main positive effect of sandblasting. Hence, the sandblasting process is designed according to its roughening effects in creating macro and/or micro-level roughness on the titanium surface (Guo et al., 2012). Our recent study indicated that sandblasting with Al2O3 powder could generate negative charges on the titanium surfaces (Guo et al., 2012). The negative surface charges on the titanium implant is believed to promote the osseointegration between the subgingival implant fixture and its adjacent bone tissues (Guo et al., 2012; Hamdan et al., 2006; Marcolonongo et al., 1998). Thus, this de novo finding might provide a new explanation to the supposed positive effects of sandblasting on osseointegration to titanium surface. This discovery has the potential to lead to optimized sandblasting techniques and, consequently, it might enable Ti implants to form faster, stronger and more durable clinical osseointegration. According to the results reported (Guo et al., 2012), the negative charge generated through sandblasting with Al2O3 powder may remain stable on the titanium surface for several minutes until it gradually diminishes. Before knowing how the charge retention is possible and what clinical applications would be feasible, it is of great interest to find out the origin of the charges. Two possible mechanisms for this charge are considered. First, the generated charges come from the titanium material itself, e.g., sandblasting removes the various titanium oxide layers from the titanium (or Ti alloy) surface, exposing electrons from the atoms of the Ti (or Ti alloy), which are detected as a negative surface charge. The removed TiO2 (Ti oxides) layer instantly reforms, which then covers up the negative charges of the exposed electrons, and could gradually reduce the negative charge detected on the titanium (alloy) surface. The charges generated on the titanium surface with this above described mechanism should always be negative, regardless of the weight and electro-negativities of blasting materials. Second, the electric charge is generated from the contact of the titanium material and the blasting powder particles. Such contact gives rise to the triboelectric effect, which is the phenomenon involving charge exchange whenever two arbitrary surfaces come into close contact. After the contact disconnects a net charge will remain on each of the surfaces (Lowell and Rose-Innes, 1980). According to the properties of the triboelectric effect certain types of blasting grits would generate a negative charge on the titanium surface, whereas some would generate a positive charge. Therefore, by observing the polarity of the charge generated by blasting grits with different positions in the so-called triboelectric series (Fig. 1, explained further in Section 4), the possible mechanism of charge generation on Ti surface by sandblasting can be deducted.
2.
Materials and methods
2.1.
Titanium plates
Fig. 1 – Example of the triboelectric series: a relative scale.
Together 60 c.p. 2 titanium plates, which were 1 mm in thickness and 15 mm in both length and width were used. All plates were pretreated by polishing with abrasive papers in the sequence of the grades 220, 320, 550 and 1000. Next
they were ultrasonically cleansed by acetone for 15 min and dried in an incubator at 37 1C overnight. The 60 Ti plates were randomly divided into 6 experimental study groups with 10 plates in each.
journal of the mechanical behavior of biomedical materials 32 (2014) 145 –154
2.2.
Sandblasting
Three different types of blasting powders in 2 grit sizes each were used in this study. They were: (a) two alumina (Al2O3) powders (Renfert GmbH, Germany) with the average diameters of 110 mm and 50 mm, (b) 45–75 mm and 150–300 mm, glass beads (Langfang Olan Glass Beads Co., Ltd, China) (a claimed SiO2 content472%), and (c) 250 mm and 44 mm aluminum powder (Yee Lee Industrial Chemical Ltd. Hong Kong). Fig. 2 shows the images obtained using a scanning electron microscope (SEM) to characterize these materials. Each type of grits was used to blast the Ti surfaces. All the sandblasting was carried out within a sandblasting chamber (SMC Corp., Tokyo, Japan). During sandblasting, the tip of the blasting pen was kept perpendicularly to the Ti plate. The distance between plate and the blasting pen was kept constantly fixed at 10 mm, with the blasting time 15 s and at a constant pressure of 3.4 bar.
2.3.
147
Static voltage measurement
An electrostatic meter (IZH10, SMC Corporation, Tokyo, Japan) was used to measure the static voltage on the titanium plate surfaces before and after sandblasting. The plates were taken out from sandblasting chamber immediately after grit blasting and the initial value of the static voltage was reported 10 s after sandblasting. After that, a series of measurement was taken at every 10 s until the static voltage on titanium surface reached the level before sandblasting for each plate specimen. The minimum display unit of the voltmeter was 0.01 kV. Some environmental factors of the experiment including the static voltage of titanium surface before sandblasting, the relative humidity and temperature are shown in Table 1. The differences in these environmental variables among different sets of experiments were interpreted negligible.
Fig. 2 – SEM micrographs of blasting materials. (a) Al2O3 powder with the diameter 110 lm ( 200); (b) Al2O3 powder with the diameter 50 lm ( 200); (c) glass beads with the diameter 150–300 lm ( 200); (d) glass beads with the diameter 45–75 lm ( 200); (e) aluminum grits with the diameter250 lm ( 100); (f) aluminum grits with the diameter 44 lm ( 200).
148
2.4.
journal of the mechanical behavior of biomedical materials 32 (2014) 145 –154
Surface roughness measurement
Roughness average (Ra), also known as the arithmetic mean roughness value (Lou et al., 1998), was measured throughout the study. Ra is the arithmetic average deviation from the mean line, obtained by the equation Z 1 L Ra ¼ ð1Þ YðxÞ dx L 0 where L is the sampling length and Y is the ordinate of the profile curve (Lou et al., 1998). The Ra values of titanium plate surface before and after sandblasting were measured. The central part of the blasted area of the Ti plate was measured by using a Surtronic 3þ (Taylor Hobson Ltd. Leicester, UK) device. L was set to 0.8 mm. The Ra values were directly measured and calculated by the device. Each plate was measured once, and the average Ra value for each study group of plates was calculated and reported.
2.5.
3.
Results
3.1.
Static charge generation
The static voltage was detected on titanium plates after sandblasting with each of the 6 groups of the powder materials. Table 2 summarizes the mean value and standard deviation of the voltages measured on the surface of the titanium plates for every experimental group at each time interval. Fig. 3(a–f) displays the average voltages and the static voltage before sandblasting for every group, respectively. Table 3 shows the average difference between the recorded static voltage and the voltage before sandblasting. Fig. 4 demonstrates the differences for the six groups in one single plot diagram. Among the blasting materials that generated a negative charge on the titanium surface, Al2O3 grits with an average diameter 110 mm generated the highest negative charge. Al grits with an average diameter 44 mm generated the lowest charge.
Statistical analysis 3.2.
The mean values and standard deviation of the static voltages on the titanium surface were calculated at all the time intervals for each blasting material. Linear regression analysis was performed on these mean values at different time intervals to find the best fitting function of the voltage. The Excel 2007 software (Microsoft, USA) was used in this analysis.
Topographical features
SEM surface analysis was performed to analyze the titanium plates0 surface textures after sandblasting using the six different groups of blasting grits. Fig. 5 shows the SEM micrographs ( 1000 magnification) of these titanium plates. Different blasting grits led to different textures on titanium surface. Larger grits led to coarser textures on the
Table 1 – Mean static voltages, humidity and temperature of the environment during the experiments. Blasting powders
Mean static voltage before sandblasting (V) Humidity (%) Temperature (oC)
Al2O3 110 mm
Al2O3 50 mm
Glass bead 150–300 mm
Glass bead 45–75 mm
Al 250 mm
Al 44 mm
þ21 47 21.5
þ36 48 21.2
þ39 46 21.3
þ32 47 21.5
þ22 43 21.7
þ34 46.5 21.7
Table 2 – Means and standard deviations of static voltages of titanium plates after sandblasting. Time t after sandblasting (s)
10 20 30 40 50 60 70 80 90 100 110 120
Al2O3 110 mm
Al2O3 50 mm
Glass beads 150–300 mm
Glass beads 45–75 mm
Al 250 mm
Al 44 mm
Mean (V)
SD (V)
Mean (V)
SD (V)
Mean (V)
SD (V)
Mean (V)
SD (V)
Mean (V)
SD (V)
Mean (V)
SD (V)
168 80 52 31 15 7 3 þ1 þ8 þ12 þ13 þ15
75.1 38.0 23.9 17.9 11.8 11.6 8.2 5.7 11.4 10.3 9.5 8.5
33 0 þ13 þ19 þ22 þ29 þ30 þ31 þ32 þ34 þ34 –
21.1 13.3 10.6 8.8 6.3 5.7 4.7 3.2 4.2 5.2 5.2 –
46 þ1 þ21 þ29 þ30 þ36 þ38 – – – – –
28.0 3.2 8.8 3.2 4.7 5.2 6.3 – – – – –
þ329 þ170 þ132 þ102 þ87 þ84 þ79 þ70 þ66 þ63 þ63 þ63
41.8 25.8 27.4 17.5 14.2 14.3 12.9 8.2 7.0 8.2 8.2 8.2
45 14 þ2 þ9 þ15 þ18 – – – – – –
37.2 19.0 14.0 9.9 10.8 9.2 – – – – – –
5 þ13 þ24 þ27 þ31 þ31 þ32 þ33 – – – –
13.5 10.6 11.7 12.5 7.4 7.4 6.3 4.8 – – – –
journal of the mechanical behavior of biomedical materials 32 (2014) 145 –154
149
Fig. 3 – Charge generation after sandblasting by (a) Al2O3 grits 110 lm; (b) Al2O3 grits 50 lm; (c) glass beads 150–300 lm; (d) glass beads 45–75 lm; (e) Al grits 250 lm; (f) Al grits 44 lm. Table 3 – Differences between mean static voltage on titanium and static voltage before sandblasting for each group at every time interval. Time t after sandblasting (s)
Al2O3 110 mm (V)
Al2O3 50 mm (V)
Glass beads 150–300 mm (V)
Glass beads 45–75 mm (V)
Al 250 mm (V)
Al 44 mm (V)
10 20 30 40 50 60 70 80 90 100 110 120
189 101 73 52 36 28 24 20 13 9 8 6
69 36 23 17 14 7 6 5 4 2 2 –
85 38 18 10 9 3 1 – – – – –
þ297 þ138 þ100 þ70 þ55 þ52 þ47 þ38 þ34 þ31 þ31 þ31
67 36 20 13 7 4 – – – – – –
39 21 10 7 3 3 2 1 – – – –
150
journal of the mechanical behavior of biomedical materials 32 (2014) 145 –154
Fig. 4 – Differences between the mean static voltage on titanium and the static voltage before sandblasting for each group at every time interval used.
titanium surface compared to smaller grits of the same material. The above observations were confirmed by the average Ra values for each group, listed in Table 4. According to the results, various blasting materials led to different levels of roughness of the titanium surface. Larger blasting grits led to higher surface roughness than smaller ones of the same material.
3.3.
Statistical analysis
As shown in Figs. 3 and 4, after sandblasting the static voltage of titanium surface as a function of time followed an exponential decay to the static voltage level before sandblasting. This pattern matched the electrostatic charge decay curve, which can be expressed by the following exponential decay equations (Malave-López and Peleg, 1985) t ð2Þ QðtÞ ¼ Qð0Þexp τ where Q(t) is amount of charge at time t, Q(0) is the amount of charge at time 0, i.e., immediately after sandblasting was stopped. τ is a constant corresponding to the rate of decay. It is noteworthy that the amount of charge on the titanium surface was directly proportional to its static voltage. Let V(t) be the voltage on the titanium surface at time t, then we have t ð3Þ VðtÞ ¼ Vð0Þexp τ By taking the natural logarithm on both sides of Eq. (3), we obtain
ln½VðtÞ ¼ ln½Vð0Þ þ
t τ
ð4Þ
The goal of the regression was thus to enumerate the value of V(0) and τ for each of the sandblasting materials. Here V(0) cannot be measured directly because taking manually the titanium plate out of the blasting chamber required some time, but V(0) can be derived in the plots (Fig. 6). The linear regression was deployed to find the best value of V(0) and τ for each of the sandblasting materials. In particular, at each time interval, we used as a data point the difference between the mean static voltage on the titanium surface and the static voltage before sandblasting, as shown in Table 3. Since the voltage can be either positive or negative, depending on its polarity, we took its absolute value before calculating the natural logarithm, which is only defined for positive values, as shown in Table 5. The mean values were used instead of the original values because the difference between original static voltages on titanium surface and the voltage before sandblasting often reached 0, for which natural logarithm is not defined. The times instances before the static voltage reached the level before sandblasting were used in the analysis. Excel 2007 was used to perform the linear regression for each of the blasting materials. The results are summarized as follows:
8 lnð V1 Þ ¼ 5:233741 0:02947t > > > > > lnð V2 Þ ¼ 4:341760:03557t > > > > < lnð V3 Þ ¼ 4:924412 0:0618t ln V 4 ¼ 5:395898 0:02196t > > > > > > lnð V5 Þ ¼ 4:723743 0:05553t > > > : lnð V Þ ¼ 4:390945 0:07093t 6
ð5Þ
journal of the mechanical behavior of biomedical materials 32 (2014) 145 –154
151
Fig. 5 – SEM micrographs ( 1000) of the titanium plates sandblasted by: (a) Al2O3 grits with the average diameter of 110 lm; (b) Al2O3 grits with the average diameter of 50 lm; (c) glass beads with the average diameter of 150–300 lm; (d) glass beads with the average diameter of 45–75 lm; (e) aluminum grits with average diameter of 250 lm; (f) aluminum grits with the average diameter of 44 lm.
In the 6 equations above, V1–V6 are the static voltage on the titanium surface after blasting with alumina grits with the diameters of 110 mm and 50 mm, glass beads with the diameters of 150–300 mm and 45–75 mm, and aluminum grits with the diameter 250 mm and 44 mm, respectively. Hence, the static voltages on the titanium surface as a function of time t for these materials are 8 V1 ¼ expð5:233741 0:02947tÞ > > > > > V2 ¼ expð4:34176 0:03557tÞ > > > > < V3 ¼ expð4:924412 0:0618tÞ ð6Þ V4 ¼ expð5:3958980:02196tÞ > > > > > > V5 ¼ expð4:723743 0:05553tÞ > > > : V6 ¼ expð4:390945 0:07093tÞ Fig. 6 plots the linear regression results.
4.
Discussion
4.1.
Origin of negative charge
To strengthen the effects of the negative charge, it is thus crucial to find out the ways to control the amount of charge generated, and the duration they remain on the titanium surface. Toward this goal, it is important to understand why and how the charges are generated during sandblasting, and the effects of the sandblasting parameters, such as the materials used as grits on charge generation. The results showed that under the same experimental conditions, different blasting materials generated charges of various magnitude and polarity on the titanium surface. In all experiments, the charges decayed relatively rapidly (Fig. 3), until they reached a stable level after several minutes. Surprisingly,
152
journal of the mechanical behavior of biomedical materials 32 (2014) 145 –154
Table 4 – Average Ra values of titanium surface for six experimental groups.
Ra (mm) SD
Al2O3 110 mm
Al2O3 50 mm
Glass beads 150–300 mm
Glass beads 45–75 mm
Al 250 mm
Al 44 mm
0.728 0.054
0.514 0.072
0.432 0.023
0.328 0.064
0.836 0.114
0.236 0.017
Fig. 6 – The mean static voltage difference (logarithmic scale) decay as a function of the time after sandblasting by (a) Al2O3 grits 110 lm; (b) Al2O3 grits 50 lm; (c) glass beads 150–300 lm; (d) glass beads 45–75 lm; (e) Al grits 250 lm; (f) Al grits 44 lm. The straight lines are the results of the linear regression. The scaling is approximately linear as expected. blasting with glass beads with the diameter 45–75 mm generated a positive charge on the titanium surface, whilst all the other blasting materials induced a negative charge. Therefore, the triboelectric effect is the most likely reason for the charges generated on the titanium surface. It is worth mentioning that as with any electrostatic voltage meter, the displayed reading depends on the size of the object being measured and the distance between the object and the sensor. In our experiments, all titanium plates
were identical in their size, and their distance to the sensor of the voltage meter was fixed to 10 mm.
4.2.
Titanium and the triboelectric effect
According to the theory of triboelectricity, whenever two arbitrary surfaces come into contact, they exchange charges. Once this contact is over, a net electric charge may exist on each of the surfaces (Lowell and Rose-Innes, 1980). This effect
journal of the mechanical behavior of biomedical materials 32 (2014) 145 –154
153
Table 5 – Natural logarithm of differences between mean static voltage on titanium and static voltage before sandblasting for each experimental group. Time t after sandblasting (s)
Al2O3 110 mm (V)
Al2O3 50 mm (V)
Glass beads 150–300 mm (V)
Glass beads 45–75 mm (V)
Al 250 mm (V)
Al 44 mm (V)
10 20 30 40 50 60 70 80 90 100 110 120
5.241747 4.615121 4.290459 3.951244 3.583519 3.332205 3.178054 2.995732 2.564949 2.197225 2.079442 1.791759
4.234107 3.583519 3.135494 2.833213 2.639057 1.94591 1.791759 1.609438 1.386294 0.693147 – –
4.442651 3.637586 2.890372 2.302585 2.197225 1.098612 – – – – – –
5.693732 4.927254 4.60517 4.248495 4.007333 3.951244 3.850148 3.637586 3.526361 3.433987 – –
4.204693 3.583519 2.995732 2.564949 1.94591 1.386294 – – – – – –
3.637586 2.995732 2.197225 1.791759 0.693147 – – – – – – –
may occur between conductors, semiconductors and insulators, solids and liquids, and even when the two surfaces are made of the same material (Castle, 1997). The sandblasting process involves multiple grit-to-grit, grit-to-air, and grit-totitanium interactions, during which the electric charges are exchanged between them (Bailey, 2001). Hence, after these interactions, a certain amount of charge may remain on the surface of the sandblasted titanium plates, as we observed in our current experiment. The triboelectric effect is a rather complex process. Although the triboelectric effect has been observed and reported a long time ago, understanding of this effect is still somewhat limited today. In particular, we are not aware of any systematic experiment reported on the triboelectric effect involving sandblasting of titanium materials. Based on the amount and direction of the charge transfer occurred in the triboelectric effect, different materials can be arranged in a sequence, i.e., the triboelectric series. Fig. 1 shows an example of the triboelectric series (reproduced from Diaz and Felix-Navarro, 2004), in which a material is expected to obtain a negative charge when it comes into contact with another material above it shown in the figure, and a positive charge after contacting with a material below it. The reason why these materials appear in this particular order in the triboelectric series is still poorly understood. Meanwhile, the literature contains contradicting results on the relative order of materials in the triboelectric series (Kanazawa et al., 1995). It is believed that the surface properties of materials are more likely to be relevant to the amount and polarity of the generated charges, rather than their bulk (Sickafoose et al., 2001). Note that the materials of the participating surfaces are an important factor that affects the amount and polarity of the charges generated through the triboelectric effect (Rose and Ward, 1957). For instance, the surface finish and purity of materials are also known to affect the generated charges (Rose and Ward, 1957). For reasons unknown, titanium is not included in the series in Fig. 1. In fact, after an extensive literature search, we could not find any existing results on the relative position of titanium compared with other materials in the triboelectric series. Furthermore, the vast majority of existing triboelectric
series (e.g., Diaz and Felix-Navarro, 2004) were produced through experiments that slowly rub two different solid materials against each other. We are not aware of any study on the triboelectric effect during sandblasting, which would involve short, rapid, and repeated contacts between the blasting grits, the titanium plate, and the air. Hence, to our best knowledge, no known result on the triboelectric series can predict the polarity of charges generated on the titanium plates during sandblasting. Our current study thus provides the initial ideas into this interesting phenomenon. Our experiments showed that the titanium plates always gain a considerable amount of the negative charge after sandblasting with Al2O3 grits. This suggested that titanium would be well below Al2O3 in the triboelectric series. Meanwhile, sandblasting with aluminum powder also consistently generated a negative charge on the titanium plates, though the amount of charge was lower compared to sandblasting with Al2O3 grits. This indicated that (i) titanium could be also below aluminum in the triboelectric series, and that (ii) titanium is closer to aluminum than it is to Al2O3. Since we could not find any existing triboelectric series that included Al2O3 or titanium, the suggested relative order of Al2O34 Al4Ti (sorted by the ability to acquire a positive charge) in a triboelectric series of sandblasting process is an interesting finding. The experimental results with glass beads were more complicated, since we observed both the positive and the negative charges on the titanium plates after sandblasting with glass beads. The beads were of different sizes: the glass beads with a larger size produced a negative charge on the titanium plates, whereas the smaller glass beads led to a positive charge on the titanium plates. This phenomenon suggested that factors other than the materials themselves dominated the polarity and the amount of charges generated through the sandblasting process. In addition to the size of the grits, two more possible factors were the speeds of the grits and the frequency of the contacts between the titanium plates and the blasting grits. In our experiments, the larger glass beads were blasted with lower speeds (as they are heavier) compared to the smaller beads. This said, for similar reasons, the larger beads also contacted the titanium plate
154
journal of the mechanical behavior of biomedical materials 32 (2014) 145 –154
less frequently than the smaller ones. Further investigations are required to find out the impact of each of these factors, and the exact reason why the polarity is reversed when we changed the size of the glass beads.
4.3.
The material of choice
Previous studies have reported that negative charge on the titanium dental implant materials may improve its osseointegration. In the current study, the results showed that sandblasting with aluminum powder or Al2O3 powder always generated the negative charge on the titanium surface, and the amount of charge depended on the size of the grits used. The smaller aluminum or Al2O3 grits led to a higher amount of charge and longer durations before the charge dissipated into the ambient atmosphere. With comparable grits sizes, sandblasting with Al2O3 powder yielded a higher amount of charge on the titanium surface than with aluminum powder. Sandblasting with glass beads might generate either a positive or a negative charge on the titanium surface, depending again on the grit size. The smaller grits could generate a weak negative charge, and the larger grits generated a positive charge. In general, Al2O3 was deemed to be the material of choice for generating a negative charge on the titanium surface through sandblasting, which might be an important factor for the favorable osseointegration properties of Al2O3 grit blasted titanium dental implants.
5.
Conclusion and future work
In the current study, some evidence of different polarity and the amount of charges generated on Ti by sandblasting was found. This was through a variety of grits of different materials which suggests that the triboelectric effect might have been responsible for the generation of such charges detected. The relative position of Al2O3 grits, aluminum grits and Ti on the triboelectric series of this special condition, i.e., sandblasting, was reported first time in the scientific literature. Further research work is necessary to improve the charge generation and retention. This could happen e.g., by choosing appropriate sandblasting parameters that could affect the triboelectric process, such as the shape and the hardness of the blasting grits.
Acknowledgements The authors are grateful to the Graduate School of the University of Hong Kong and the Prince Philip Dental Hospital
for their continuous support. We would also like to thank Baoji Xinlian Titanium Industry Co., Ltd (http://www.xin lianti.com) for their generous supplement of titanium materi als. We would also like thank the help from Mr Simon Lee, Mr Tony Yuen, Ms Meng Zhang, Dr Quan Liu and Dr Dan Liu in all the experiments.
r e f e r e n c e s
Al Jabbari, Y.S., Zinelis, S., Eliades, G., 2012. Effect of sandblasting conditions on alumina retention in representative dental alloys. Dent. Mater. J. 31, 249–255. Bailey, A.G., 2001. The charging of insulator surfaces. J. Electrostat. 51–52, 82–90. Castle, G.S.P., 1997. Contact charging between insulators. J. Electrostat. 40–41, 13–20. Diaz, A.F., Felix-Navarro, R.M., 2004. A semi-quantitative triboelectric series for polymeric materials: the influence of chemical structure and properties. J. Electrostat 62, 277–290. Guo, C.Y., Matinlinna, J.P., Tang, A.T.H., 2012. A novel effect of sandblasting on titanium surface: static charge generation. J. Adhes. Sci. Technol. 26, 2603–2613. Guo, C.Y., Tang, A.T.H., Matinlinna, J.P., 2012. Insights into surface treatment methods of titanium dental implants. J. Adhes. Sci. Technol. 26, 189–205. Hamdan, M., Blanco, L., Khraisat, A., Tresguerres, I.F., 2006. Influence of titanium surface charge on fibroblast adhesion. Clin. Implant Dent. Relat. Res. 8, 32–38. Kanazawa, S., Ohkubo, T., Nomoto, Y., Adachi, T., 1995. Electrification of a pipe wall during powder transport. J. Electrostat. 35, 47–54. Lou, M.J., Chen, J.C., Li, C.M., 1998. Surface roughness prediction technique for CNC end-milling. J. Ind. Technol 15, 1–6. Lowell, J., Rose-Innes, A.C., 1980. Contact electrification. Adv. Phys. 29, 947–1023. Malave-Lo´pez, J., Peleg, M., 1985. Linearization of the electrostatic charging and charge decay curves of powders. Powder Technol. 42, 217–223. Marcolonongo, M., Dinardo, N.J., Hickok, N., Tuan, R., Pourezzaei, K., Beard, R., Brennan, D., Heipp, P., Phan, T., 1998. Osteoblast attachment on biomaterials as a function of surface charge. MRS Proc. 550, 121–126. Papadopoulos, T., Tsetsekou, A., Eliades, G., 1999. Effect of aluminium oxide sandblasting on cast commercially pure titanium surface. Eur. J. Prosthod. Restor. Dent 7, 15–21. Pazos, L., Corengia, P., Svoboda, H., 2010. Effect of surface treatments on the fatigue life of titanium for biomedical application. J. Mech. Behav. Biomed. Mater 3, 416–424. Rose, G.S., Ward, S.G., 1957. Contact electrification across metal–dielectric and dielectric–dielectric interfaces. Br. J. Appl. Phys. 8, 121–126. Sickafoose, A.A., Colwell, J.E., Hora´nyi, M., Robertson, S., 2001. Experimental investigations on photoelectric and triboelectric charging of dust. J. Geophys. Res. 106, 8343–8356.
Available online at www.sciencedirect.com
www.elsevier.com/locate/jmbbm
Research Paper
Effects of different blasting materials on charge generation and decay on titanium surface after sandblasting Cecilia Yan Guo, Alexander Tin Hong Tang, James Kit Hon Tsoi, Jukka Pekka Matinlinnan Dental Materials Science, Faculty of Dentistry, The University of Hong Kong, 34 Hospital Road, Sai Ying Pun, Hong Kong, PR China
art i cle i nfo
ab st rac t
Article history:
It has been reported that sandblasting titanium with alumina (Al2O3) powder could
Received 11 September 2013
generate a negative electric charge on titanium surface. This has been proven to promote
Received in revised form
osteoblast activities and possibly osseointegration. The purpose of this pilot study was to
21 December 2013
investigate the effects of different blasting materials, in terms of the grit sizes and electro-
Accepted 28 December 2013
negativity, on the generation of a negative charge on the titanium surface. The aim was
Available online 6 January 2014
also to make use of these results to deduct the underlying mechanism of charge generation
Keywords:
by sandblasting.
Titanium Dental implants Sandblasting
Together 60 c.p. 2 titanium plates were machine-cut and polished for sandblasting, and divided into 6 groups with 10 plates in each. Every plate in the study groups was sandblasted with one of the following 6 powder materials: 110 mm Al2O3 grits, 50 mm
Electric charge
Al2O3 grits, 150–300 mm glass beads, 45–75 mm glass beads, 250 mm Al powder and 44 mm Al
Alumina powder
powder. The static voltage on the surface of every titanium plate was measured
Glass beads
immediately after sandblasting. The static voltages of the titanium plates were recorded
Aluminum powder
and processed using statistical analysis. The results suggested that only sandblasting with 45–75 mm glass beads generated a positive charge on titanium, while using all other blasting materials lead to a negative charge. Furthermore, blasting grits of the same powder material but of different sizes might lead to different amount and polarity of the charges. This triboelectric effect is likely to be the main mechanism for charge generation through sandblasting. & 2014 Elsevier Ltd. All rights reserved.
1.
Introduction
Sandblasting is a commonly used surface treatment method for titanium dental implants to improve on their bone-toimplant bonding. Previous studies have mostly focused on n
Corresponding author. Tel.: þ852 285 90380; fax: þ852 254 89464. E-mail address: [email protected] (J.P. Matinlinna).
1751-6161/$ - see front matter & 2014 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.jmbbm.2013.12.026
the mechanical impacts of sandblasting on the titanium surface, such as the created roughness (Papadopoulos et al., 1999), residual stress, subsurface microstructure (Pazos et al., 2010) and eventual contamination (Al Jabbari et al., 2012). Among these effects, modifying the surface roughness of
146
journal of the mechanical behavior of biomedical materials 32 (2014) 145 –154
titanium and its alloys has been considered the main positive effect of sandblasting. Hence, the sandblasting process is designed according to its roughening effects in creating macro and/or micro-level roughness on the titanium surface (Guo et al., 2012). Our recent study indicated that sandblasting with Al2O3 powder could generate negative charges on the titanium surfaces (Guo et al., 2012). The negative surface charges on the titanium implant is believed to promote the osseointegration between the subgingival implant fixture and its adjacent bone tissues (Guo et al., 2012; Hamdan et al., 2006; Marcolonongo et al., 1998). Thus, this de novo finding might provide a new explanation to the supposed positive effects of sandblasting on osseointegration to titanium surface. This discovery has the potential to lead to optimized sandblasting techniques and, consequently, it might enable Ti implants to form faster, stronger and more durable clinical osseointegration. According to the results reported (Guo et al., 2012), the negative charge generated through sandblasting with Al2O3 powder may remain stable on the titanium surface for several minutes until it gradually diminishes. Before knowing how the charge retention is possible and what clinical applications would be feasible, it is of great interest to find out the origin of the charges. Two possible mechanisms for this charge are considered. First, the generated charges come from the titanium material itself, e.g., sandblasting removes the various titanium oxide layers from the titanium (or Ti alloy) surface, exposing electrons from the atoms of the Ti (or Ti alloy), which are detected as a negative surface charge. The removed TiO2 (Ti oxides) layer instantly reforms, which then covers up the negative charges of the exposed electrons, and could gradually reduce the negative charge detected on the titanium (alloy) surface. The charges generated on the titanium surface with this above described mechanism should always be negative, regardless of the weight and electro-negativities of blasting materials. Second, the electric charge is generated from the contact of the titanium material and the blasting powder particles. Such contact gives rise to the triboelectric effect, which is the phenomenon involving charge exchange whenever two arbitrary surfaces come into close contact. After the contact disconnects a net charge will remain on each of the surfaces (Lowell and Rose-Innes, 1980). According to the properties of the triboelectric effect certain types of blasting grits would generate a negative charge on the titanium surface, whereas some would generate a positive charge. Therefore, by observing the polarity of the charge generated by blasting grits with different positions in the so-called triboelectric series (Fig. 1, explained further in Section 4), the possible mechanism of charge generation on Ti surface by sandblasting can be deducted.
2.
Materials and methods
2.1.
Titanium plates
Fig. 1 – Example of the triboelectric series: a relative scale.
Together 60 c.p. 2 titanium plates, which were 1 mm in thickness and 15 mm in both length and width were used. All plates were pretreated by polishing with abrasive papers in the sequence of the grades 220, 320, 550 and 1000. Next
they were ultrasonically cleansed by acetone for 15 min and dried in an incubator at 37 1C overnight. The 60 Ti plates were randomly divided into 6 experimental study groups with 10 plates in each.
journal of the mechanical behavior of biomedical materials 32 (2014) 145 –154
2.2.
Sandblasting
Three different types of blasting powders in 2 grit sizes each were used in this study. They were: (a) two alumina (Al2O3) powders (Renfert GmbH, Germany) with the average diameters of 110 mm and 50 mm, (b) 45–75 mm and 150–300 mm, glass beads (Langfang Olan Glass Beads Co., Ltd, China) (a claimed SiO2 content472%), and (c) 250 mm and 44 mm aluminum powder (Yee Lee Industrial Chemical Ltd. Hong Kong). Fig. 2 shows the images obtained using a scanning electron microscope (SEM) to characterize these materials. Each type of grits was used to blast the Ti surfaces. All the sandblasting was carried out within a sandblasting chamber (SMC Corp., Tokyo, Japan). During sandblasting, the tip of the blasting pen was kept perpendicularly to the Ti plate. The distance between plate and the blasting pen was kept constantly fixed at 10 mm, with the blasting time 15 s and at a constant pressure of 3.4 bar.
2.3.
147
Static voltage measurement
An electrostatic meter (IZH10, SMC Corporation, Tokyo, Japan) was used to measure the static voltage on the titanium plate surfaces before and after sandblasting. The plates were taken out from sandblasting chamber immediately after grit blasting and the initial value of the static voltage was reported 10 s after sandblasting. After that, a series of measurement was taken at every 10 s until the static voltage on titanium surface reached the level before sandblasting for each plate specimen. The minimum display unit of the voltmeter was 0.01 kV. Some environmental factors of the experiment including the static voltage of titanium surface before sandblasting, the relative humidity and temperature are shown in Table 1. The differences in these environmental variables among different sets of experiments were interpreted negligible.
Fig. 2 – SEM micrographs of blasting materials. (a) Al2O3 powder with the diameter 110 lm ( 200); (b) Al2O3 powder with the diameter 50 lm ( 200); (c) glass beads with the diameter 150–300 lm ( 200); (d) glass beads with the diameter 45–75 lm ( 200); (e) aluminum grits with the diameter250 lm ( 100); (f) aluminum grits with the diameter 44 lm ( 200).
148
2.4.
journal of the mechanical behavior of biomedical materials 32 (2014) 145 –154
Surface roughness measurement
Roughness average (Ra), also known as the arithmetic mean roughness value (Lou et al., 1998), was measured throughout the study. Ra is the arithmetic average deviation from the mean line, obtained by the equation Z 1 L Ra ¼ ð1Þ YðxÞ dx L 0 where L is the sampling length and Y is the ordinate of the profile curve (Lou et al., 1998). The Ra values of titanium plate surface before and after sandblasting were measured. The central part of the blasted area of the Ti plate was measured by using a Surtronic 3þ (Taylor Hobson Ltd. Leicester, UK) device. L was set to 0.8 mm. The Ra values were directly measured and calculated by the device. Each plate was measured once, and the average Ra value for each study group of plates was calculated and reported.
2.5.
3.
Results
3.1.
Static charge generation
The static voltage was detected on titanium plates after sandblasting with each of the 6 groups of the powder materials. Table 2 summarizes the mean value and standard deviation of the voltages measured on the surface of the titanium plates for every experimental group at each time interval. Fig. 3(a–f) displays the average voltages and the static voltage before sandblasting for every group, respectively. Table 3 shows the average difference between the recorded static voltage and the voltage before sandblasting. Fig. 4 demonstrates the differences for the six groups in one single plot diagram. Among the blasting materials that generated a negative charge on the titanium surface, Al2O3 grits with an average diameter 110 mm generated the highest negative charge. Al grits with an average diameter 44 mm generated the lowest charge.
Statistical analysis 3.2.
The mean values and standard deviation of the static voltages on the titanium surface were calculated at all the time intervals for each blasting material. Linear regression analysis was performed on these mean values at different time intervals to find the best fitting function of the voltage. The Excel 2007 software (Microsoft, USA) was used in this analysis.
Topographical features
SEM surface analysis was performed to analyze the titanium plates0 surface textures after sandblasting using the six different groups of blasting grits. Fig. 5 shows the SEM micrographs ( 1000 magnification) of these titanium plates. Different blasting grits led to different textures on titanium surface. Larger grits led to coarser textures on the
Table 1 – Mean static voltages, humidity and temperature of the environment during the experiments. Blasting powders
Mean static voltage before sandblasting (V) Humidity (%) Temperature (oC)
Al2O3 110 mm
Al2O3 50 mm
Glass bead 150–300 mm
Glass bead 45–75 mm
Al 250 mm
Al 44 mm
þ21 47 21.5
þ36 48 21.2
þ39 46 21.3
þ32 47 21.5
þ22 43 21.7
þ34 46.5 21.7
Table 2 – Means and standard deviations of static voltages of titanium plates after sandblasting. Time t after sandblasting (s)
10 20 30 40 50 60 70 80 90 100 110 120
Al2O3 110 mm
Al2O3 50 mm
Glass beads 150–300 mm
Glass beads 45–75 mm
Al 250 mm
Al 44 mm
Mean (V)
SD (V)
Mean (V)
SD (V)
Mean (V)
SD (V)
Mean (V)
SD (V)
Mean (V)
SD (V)
Mean (V)
SD (V)
168 80 52 31 15 7 3 þ1 þ8 þ12 þ13 þ15
75.1 38.0 23.9 17.9 11.8 11.6 8.2 5.7 11.4 10.3 9.5 8.5
33 0 þ13 þ19 þ22 þ29 þ30 þ31 þ32 þ34 þ34 –
21.1 13.3 10.6 8.8 6.3 5.7 4.7 3.2 4.2 5.2 5.2 –
46 þ1 þ21 þ29 þ30 þ36 þ38 – – – – –
28.0 3.2 8.8 3.2 4.7 5.2 6.3 – – – – –
þ329 þ170 þ132 þ102 þ87 þ84 þ79 þ70 þ66 þ63 þ63 þ63
41.8 25.8 27.4 17.5 14.2 14.3 12.9 8.2 7.0 8.2 8.2 8.2
45 14 þ2 þ9 þ15 þ18 – – – – – –
37.2 19.0 14.0 9.9 10.8 9.2 – – – – – –
5 þ13 þ24 þ27 þ31 þ31 þ32 þ33 – – – –
13.5 10.6 11.7 12.5 7.4 7.4 6.3 4.8 – – – –
journal of the mechanical behavior of biomedical materials 32 (2014) 145 –154
149
Fig. 3 – Charge generation after sandblasting by (a) Al2O3 grits 110 lm; (b) Al2O3 grits 50 lm; (c) glass beads 150–300 lm; (d) glass beads 45–75 lm; (e) Al grits 250 lm; (f) Al grits 44 lm. Table 3 – Differences between mean static voltage on titanium and static voltage before sandblasting for each group at every time interval. Time t after sandblasting (s)
Al2O3 110 mm (V)
Al2O3 50 mm (V)
Glass beads 150–300 mm (V)
Glass beads 45–75 mm (V)
Al 250 mm (V)
Al 44 mm (V)
10 20 30 40 50 60 70 80 90 100 110 120
189 101 73 52 36 28 24 20 13 9 8 6
69 36 23 17 14 7 6 5 4 2 2 –
85 38 18 10 9 3 1 – – – – –
þ297 þ138 þ100 þ70 þ55 þ52 þ47 þ38 þ34 þ31 þ31 þ31
67 36 20 13 7 4 – – – – – –
39 21 10 7 3 3 2 1 – – – –
150
journal of the mechanical behavior of biomedical materials 32 (2014) 145 –154
Fig. 4 – Differences between the mean static voltage on titanium and the static voltage before sandblasting for each group at every time interval used.
titanium surface compared to smaller grits of the same material. The above observations were confirmed by the average Ra values for each group, listed in Table 4. According to the results, various blasting materials led to different levels of roughness of the titanium surface. Larger blasting grits led to higher surface roughness than smaller ones of the same material.
3.3.
Statistical analysis
As shown in Figs. 3 and 4, after sandblasting the static voltage of titanium surface as a function of time followed an exponential decay to the static voltage level before sandblasting. This pattern matched the electrostatic charge decay curve, which can be expressed by the following exponential decay equations (Malave-López and Peleg, 1985) t ð2Þ QðtÞ ¼ Qð0Þexp τ where Q(t) is amount of charge at time t, Q(0) is the amount of charge at time 0, i.e., immediately after sandblasting was stopped. τ is a constant corresponding to the rate of decay. It is noteworthy that the amount of charge on the titanium surface was directly proportional to its static voltage. Let V(t) be the voltage on the titanium surface at time t, then we have t ð3Þ VðtÞ ¼ Vð0Þexp τ By taking the natural logarithm on both sides of Eq. (3), we obtain
ln½VðtÞ ¼ ln½Vð0Þ þ
t τ
ð4Þ
The goal of the regression was thus to enumerate the value of V(0) and τ for each of the sandblasting materials. Here V(0) cannot be measured directly because taking manually the titanium plate out of the blasting chamber required some time, but V(0) can be derived in the plots (Fig. 6). The linear regression was deployed to find the best value of V(0) and τ for each of the sandblasting materials. In particular, at each time interval, we used as a data point the difference between the mean static voltage on the titanium surface and the static voltage before sandblasting, as shown in Table 3. Since the voltage can be either positive or negative, depending on its polarity, we took its absolute value before calculating the natural logarithm, which is only defined for positive values, as shown in Table 5. The mean values were used instead of the original values because the difference between original static voltages on titanium surface and the voltage before sandblasting often reached 0, for which natural logarithm is not defined. The times instances before the static voltage reached the level before sandblasting were used in the analysis. Excel 2007 was used to perform the linear regression for each of the blasting materials. The results are summarized as follows:
8 lnð V1 Þ ¼ 5:233741 0:02947t > > > > > lnð V2 Þ ¼ 4:341760:03557t > > > > < lnð V3 Þ ¼ 4:924412 0:0618t ln V 4 ¼ 5:395898 0:02196t > > > > > > lnð V5 Þ ¼ 4:723743 0:05553t > > > : lnð V Þ ¼ 4:390945 0:07093t 6
ð5Þ
journal of the mechanical behavior of biomedical materials 32 (2014) 145 –154
151
Fig. 5 – SEM micrographs ( 1000) of the titanium plates sandblasted by: (a) Al2O3 grits with the average diameter of 110 lm; (b) Al2O3 grits with the average diameter of 50 lm; (c) glass beads with the average diameter of 150–300 lm; (d) glass beads with the average diameter of 45–75 lm; (e) aluminum grits with average diameter of 250 lm; (f) aluminum grits with the average diameter of 44 lm.
In the 6 equations above, V1–V6 are the static voltage on the titanium surface after blasting with alumina grits with the diameters of 110 mm and 50 mm, glass beads with the diameters of 150–300 mm and 45–75 mm, and aluminum grits with the diameter 250 mm and 44 mm, respectively. Hence, the static voltages on the titanium surface as a function of time t for these materials are 8 V1 ¼ expð5:233741 0:02947tÞ > > > > > V2 ¼ expð4:34176 0:03557tÞ > > > > < V3 ¼ expð4:924412 0:0618tÞ ð6Þ V4 ¼ expð5:3958980:02196tÞ > > > > > > V5 ¼ expð4:723743 0:05553tÞ > > > : V6 ¼ expð4:390945 0:07093tÞ Fig. 6 plots the linear regression results.
4.
Discussion
4.1.
Origin of negative charge
To strengthen the effects of the negative charge, it is thus crucial to find out the ways to control the amount of charge generated, and the duration they remain on the titanium surface. Toward this goal, it is important to understand why and how the charges are generated during sandblasting, and the effects of the sandblasting parameters, such as the materials used as grits on charge generation. The results showed that under the same experimental conditions, different blasting materials generated charges of various magnitude and polarity on the titanium surface. In all experiments, the charges decayed relatively rapidly (Fig. 3), until they reached a stable level after several minutes. Surprisingly,
152
journal of the mechanical behavior of biomedical materials 32 (2014) 145 –154
Table 4 – Average Ra values of titanium surface for six experimental groups.
Ra (mm) SD
Al2O3 110 mm
Al2O3 50 mm
Glass beads 150–300 mm
Glass beads 45–75 mm
Al 250 mm
Al 44 mm
0.728 0.054
0.514 0.072
0.432 0.023
0.328 0.064
0.836 0.114
0.236 0.017
Fig. 6 – The mean static voltage difference (logarithmic scale) decay as a function of the time after sandblasting by (a) Al2O3 grits 110 lm; (b) Al2O3 grits 50 lm; (c) glass beads 150–300 lm; (d) glass beads 45–75 lm; (e) Al grits 250 lm; (f) Al grits 44 lm. The straight lines are the results of the linear regression. The scaling is approximately linear as expected. blasting with glass beads with the diameter 45–75 mm generated a positive charge on the titanium surface, whilst all the other blasting materials induced a negative charge. Therefore, the triboelectric effect is the most likely reason for the charges generated on the titanium surface. It is worth mentioning that as with any electrostatic voltage meter, the displayed reading depends on the size of the object being measured and the distance between the object and the sensor. In our experiments, all titanium plates
were identical in their size, and their distance to the sensor of the voltage meter was fixed to 10 mm.
4.2.
Titanium and the triboelectric effect
According to the theory of triboelectricity, whenever two arbitrary surfaces come into contact, they exchange charges. Once this contact is over, a net electric charge may exist on each of the surfaces (Lowell and Rose-Innes, 1980). This effect
journal of the mechanical behavior of biomedical materials 32 (2014) 145 –154
153
Table 5 – Natural logarithm of differences between mean static voltage on titanium and static voltage before sandblasting for each experimental group. Time t after sandblasting (s)
Al2O3 110 mm (V)
Al2O3 50 mm (V)
Glass beads 150–300 mm (V)
Glass beads 45–75 mm (V)
Al 250 mm (V)
Al 44 mm (V)
10 20 30 40 50 60 70 80 90 100 110 120
5.241747 4.615121 4.290459 3.951244 3.583519 3.332205 3.178054 2.995732 2.564949 2.197225 2.079442 1.791759
4.234107 3.583519 3.135494 2.833213 2.639057 1.94591 1.791759 1.609438 1.386294 0.693147 – –
4.442651 3.637586 2.890372 2.302585 2.197225 1.098612 – – – – – –
5.693732 4.927254 4.60517 4.248495 4.007333 3.951244 3.850148 3.637586 3.526361 3.433987 – –
4.204693 3.583519 2.995732 2.564949 1.94591 1.386294 – – – – – –
3.637586 2.995732 2.197225 1.791759 0.693147 – – – – – – –
may occur between conductors, semiconductors and insulators, solids and liquids, and even when the two surfaces are made of the same material (Castle, 1997). The sandblasting process involves multiple grit-to-grit, grit-to-air, and grit-totitanium interactions, during which the electric charges are exchanged between them (Bailey, 2001). Hence, after these interactions, a certain amount of charge may remain on the surface of the sandblasted titanium plates, as we observed in our current experiment. The triboelectric effect is a rather complex process. Although the triboelectric effect has been observed and reported a long time ago, understanding of this effect is still somewhat limited today. In particular, we are not aware of any systematic experiment reported on the triboelectric effect involving sandblasting of titanium materials. Based on the amount and direction of the charge transfer occurred in the triboelectric effect, different materials can be arranged in a sequence, i.e., the triboelectric series. Fig. 1 shows an example of the triboelectric series (reproduced from Diaz and Felix-Navarro, 2004), in which a material is expected to obtain a negative charge when it comes into contact with another material above it shown in the figure, and a positive charge after contacting with a material below it. The reason why these materials appear in this particular order in the triboelectric series is still poorly understood. Meanwhile, the literature contains contradicting results on the relative order of materials in the triboelectric series (Kanazawa et al., 1995). It is believed that the surface properties of materials are more likely to be relevant to the amount and polarity of the generated charges, rather than their bulk (Sickafoose et al., 2001). Note that the materials of the participating surfaces are an important factor that affects the amount and polarity of the charges generated through the triboelectric effect (Rose and Ward, 1957). For instance, the surface finish and purity of materials are also known to affect the generated charges (Rose and Ward, 1957). For reasons unknown, titanium is not included in the series in Fig. 1. In fact, after an extensive literature search, we could not find any existing results on the relative position of titanium compared with other materials in the triboelectric series. Furthermore, the vast majority of existing triboelectric
series (e.g., Diaz and Felix-Navarro, 2004) were produced through experiments that slowly rub two different solid materials against each other. We are not aware of any study on the triboelectric effect during sandblasting, which would involve short, rapid, and repeated contacts between the blasting grits, the titanium plate, and the air. Hence, to our best knowledge, no known result on the triboelectric series can predict the polarity of charges generated on the titanium plates during sandblasting. Our current study thus provides the initial ideas into this interesting phenomenon. Our experiments showed that the titanium plates always gain a considerable amount of the negative charge after sandblasting with Al2O3 grits. This suggested that titanium would be well below Al2O3 in the triboelectric series. Meanwhile, sandblasting with aluminum powder also consistently generated a negative charge on the titanium plates, though the amount of charge was lower compared to sandblasting with Al2O3 grits. This indicated that (i) titanium could be also below aluminum in the triboelectric series, and that (ii) titanium is closer to aluminum than it is to Al2O3. Since we could not find any existing triboelectric series that included Al2O3 or titanium, the suggested relative order of Al2O34 Al4Ti (sorted by the ability to acquire a positive charge) in a triboelectric series of sandblasting process is an interesting finding. The experimental results with glass beads were more complicated, since we observed both the positive and the negative charges on the titanium plates after sandblasting with glass beads. The beads were of different sizes: the glass beads with a larger size produced a negative charge on the titanium plates, whereas the smaller glass beads led to a positive charge on the titanium plates. This phenomenon suggested that factors other than the materials themselves dominated the polarity and the amount of charges generated through the sandblasting process. In addition to the size of the grits, two more possible factors were the speeds of the grits and the frequency of the contacts between the titanium plates and the blasting grits. In our experiments, the larger glass beads were blasted with lower speeds (as they are heavier) compared to the smaller beads. This said, for similar reasons, the larger beads also contacted the titanium plate
154
journal of the mechanical behavior of biomedical materials 32 (2014) 145 –154
less frequently than the smaller ones. Further investigations are required to find out the impact of each of these factors, and the exact reason why the polarity is reversed when we changed the size of the glass beads.
4.3.
The material of choice
Previous studies have reported that negative charge on the titanium dental implant materials may improve its osseointegration. In the current study, the results showed that sandblasting with aluminum powder or Al2O3 powder always generated the negative charge on the titanium surface, and the amount of charge depended on the size of the grits used. The smaller aluminum or Al2O3 grits led to a higher amount of charge and longer durations before the charge dissipated into the ambient atmosphere. With comparable grits sizes, sandblasting with Al2O3 powder yielded a higher amount of charge on the titanium surface than with aluminum powder. Sandblasting with glass beads might generate either a positive or a negative charge on the titanium surface, depending again on the grit size. The smaller grits could generate a weak negative charge, and the larger grits generated a positive charge. In general, Al2O3 was deemed to be the material of choice for generating a negative charge on the titanium surface through sandblasting, which might be an important factor for the favorable osseointegration properties of Al2O3 grit blasted titanium dental implants.
5.
Conclusion and future work
In the current study, some evidence of different polarity and the amount of charges generated on Ti by sandblasting was found. This was through a variety of grits of different materials which suggests that the triboelectric effect might have been responsible for the generation of such charges detected. The relative position of Al2O3 grits, aluminum grits and Ti on the triboelectric series of this special condition, i.e., sandblasting, was reported first time in the scientific literature. Further research work is necessary to improve the charge generation and retention. This could happen e.g., by choosing appropriate sandblasting parameters that could affect the triboelectric process, such as the shape and the hardness of the blasting grits.
Acknowledgements The authors are grateful to the Graduate School of the University of Hong Kong and the Prince Philip Dental Hospital
for their continuous support. We would also like to thank Baoji Xinlian Titanium Industry Co., Ltd (http://www.xin lianti.com) for their generous supplement of titanium materi als. We would also like thank the help from Mr Simon Lee, Mr Tony Yuen, Ms Meng Zhang, Dr Quan Liu and Dr Dan Liu in all the experiments.
r e f e r e n c e s
Al Jabbari, Y.S., Zinelis, S., Eliades, G., 2012. Effect of sandblasting conditions on alumina retention in representative dental alloys. Dent. Mater. J. 31, 249–255. Bailey, A.G., 2001. The charging of insulator surfaces. J. Electrostat. 51–52, 82–90. Castle, G.S.P., 1997. Contact charging between insulators. J. Electrostat. 40–41, 13–20. Diaz, A.F., Felix-Navarro, R.M., 2004. A semi-quantitative triboelectric series for polymeric materials: the influence of chemical structure and properties. J. Electrostat 62, 277–290. Guo, C.Y., Matinlinna, J.P., Tang, A.T.H., 2012. A novel effect of sandblasting on titanium surface: static charge generation. J. Adhes. Sci. Technol. 26, 2603–2613. Guo, C.Y., Tang, A.T.H., Matinlinna, J.P., 2012. Insights into surface treatment methods of titanium dental implants. J. Adhes. Sci. Technol. 26, 189–205. Hamdan, M., Blanco, L., Khraisat, A., Tresguerres, I.F., 2006. Influence of titanium surface charge on fibroblast adhesion. Clin. Implant Dent. Relat. Res. 8, 32–38. Kanazawa, S., Ohkubo, T., Nomoto, Y., Adachi, T., 1995. Electrification of a pipe wall during powder transport. J. Electrostat. 35, 47–54. Lou, M.J., Chen, J.C., Li, C.M., 1998. Surface roughness prediction technique for CNC end-milling. J. Ind. Technol 15, 1–6. Lowell, J., Rose-Innes, A.C., 1980. Contact electrification. Adv. Phys. 29, 947–1023. Malave-Lo´pez, J., Peleg, M., 1985. Linearization of the electrostatic charging and charge decay curves of powders. Powder Technol. 42, 217–223. Marcolonongo, M., Dinardo, N.J., Hickok, N., Tuan, R., Pourezzaei, K., Beard, R., Brennan, D., Heipp, P., Phan, T., 1998. Osteoblast attachment on biomaterials as a function of surface charge. MRS Proc. 550, 121–126. Papadopoulos, T., Tsetsekou, A., Eliades, G., 1999. Effect of aluminium oxide sandblasting on cast commercially pure titanium surface. Eur. J. Prosthod. Restor. Dent 7, 15–21. Pazos, L., Corengia, P., Svoboda, H., 2010. Effect of surface treatments on the fatigue life of titanium for biomedical application. J. Mech. Behav. Biomed. Mater 3, 416–424. Rose, G.S., Ward, S.G., 1957. Contact electrification across metal–dielectric and dielectric–dielectric interfaces. Br. J. Appl. Phys. 8, 121–126. Sickafoose, A.A., Colwell, J.E., Hora´nyi, M., Robertson, S., 2001. Experimental investigations on photoelectric and triboelectric charging of dust. J. Geophys. Res. 106, 8343–8356.
