Processing and Characterization of Hierarchical TiO2

Download 7 Mb.
Size7 Mb.
c:\documents and settings\christi mclinn\my documents\downloads\m logo transparent background.gifur.2col.v2.epsc:\documents and settings\christi mclinn\my documents\downloads\nsf_logo.jpg

Processing and Characterization of Hierarchical TiO2 Coatings on Ti Implant Materials: Cold Spray and Biological Response

Prepared by: Christine McLinn

Faculty Advisor: Dr. Grant Crawford

Research Experience for Undergraduates

National Science Foundation REU Site Award #1157074

Summer 2012

Department of Materials and Metallurgical Engineering

South Dakota School of Mines and Technology

Rapid City, South Dakota

Table of Contents

Abstract------------------------------------------------------------------------ 3

Introduction------------------------------------------------------------------- 3

Experimental Procedure




Results------------------------------------------------------------------------ 10

Discussion-------------------------------------------------------------------- 19





Titanium is inert and therefore corrosion resistant making it an ideal metal to use for medical devices. Although it is already commonly used for orthopedic implants advancements, must be made to improve implant lifetimes. In this work we report on the use of a microstructure created by cold spray with a nanostructure of TiO2 nanotubes on top to improve osseointegration between titanium implants and patient’s bones. After a week of cell culture alizarin red staining indicated a greater amount of calcium on samples containing both the micro and nanostructure compared to those samples that had a single structure or plain titanium. While further biological testing is necessary, these results indicate that hierarchical surface modifications on titanium may be a viable approach to increase osseointegration for orthopedic implants.


Titanium is a metal often used in biomedical implants, however there is a need for improved integration of these implants into the body. A major problem that frequently arises with orthopedic implants is stress shielding [1]. Since implants are strong and stiff and tend to be placed in locations that bear a lot of load, bone around them often resorbs which can cause loosening of the implants leading to discomfort and eventual failure. As the world population ages more people need implants like knee and hip replacements and the current implants simply do not last long enough [2]. To improve the life of orthopedic implants one option is to modify the surface of the implants to promote osseointegration. A better fixation of implants to bone may be able to increase the lifespan of the implant while reducing the need for revision surgeries.

To address this need for advancement in implant technology, it is proposed to use a combination of surface modifications in an attempt to create better integration. This project uses anodic oxidation to create a TiO2 nanotube surface coating on top of cold spray and attempts to determine whether these structures will enhance biocompatibility of titanium implants. Previous work has shown that nanotubes can increase bioactivity of biomedical implants and that laser deposition with various titanium alloys also improves osseointegration [2,3,4]. In fact TiO2 nanotubes have been shown to increase the rate of osteoblast growth by up to 300-400% which is a significant amount [5]. However, the combination of a nanostructure with the microstructure created using cold spray is untested. The goal of this project is to determine whether the mixture of different sized surface structures will cause enhanced osseointegration compared to what is seen with current orthopedic implants. Cell culture will be used to determine the amount of osseointegration by looking at the level of calcification and alkaline phosphatase activity.

Experimental Procedure


  • titanium samples .74mm thick with purity of 99.4+% (annealed)

  • titanium wire .25mm thick with purity of 99.7%

  • 40 micron titanium particles for cold spray

  • Platinum electrode


  • Electrochemical cell (250 mL plastic beaker)

  • Buehler MetaServ 250 Grinder-Polisher

  • FESEM: Zeiss Supra 40 VP

  • Cell culture equipment: Biological Safety Cabinet (Baker SterilGard Model SG-600), Centrifuge (Sorvall RT6000B), Forma Scientific CO2 water jacketed incubator, Autoclave (Pelton & Crane MagnaClave), Thermolyne Locator 4 cryo biological storage system

  • Cold Spray Equipment

  • Profilometer (Mahr Pocket Surf PS1)


Titanium Polishing and Preparation

Before adding surface modifications to the titanium the samples first had to be cut from a sheet down to 1.5 x 1.5 cm using a diamond saw and a hole was drilled through the top center of the sample using a size 1 center drill bit. The samples were then polished on 600 grit silicon carbide paper for a few minutes until they appeared brighter. Then the samples were turned 90 degrees from the direction they were originally held and placed on 1200 grit paper. After the titanium was polished using 1 micron alumina on a felt pad. This was stopped when there were no longer large visible scratches. The last step was to polish the sample on a felt pad using 0.3 µm alumina to remove any remaining small scratches. When there were no longer extremely apparent scratches and the sample had a mirror like finish polishing was stopped. This procedure sometimes resulted in samples that contained wave like structures that were deep scratches that could not be removed within a reasonable amount of polishing time. In order to avoid these in the future a chemical polish could be used.

After polishing the titanium sample was placed in a beaker of deionized water and sonicated for 5 minutes to remove any particles remaining on the surface. The 0.25mm thick titanium wire was cut into a 6 inch piece where a half an inch was bent. This bent portion was then placed through the hole in the sample and twisted around itself to secure the sample to the wire. The titanium wire and sample combined made up the titanium electrode or the anode of the electrochemical cell. Both electrodes were then held in inside a 250 mL plastic beaker using electrical tape. The top of the platinum and titanium electrodes were located 5.3 and 5.5 cm from the top of the beaker respectively.

Nanotube Processing

First a 150mL solution of 1 M H2SO4 was created by pouring141.7 mL of deionized water into a plastic beaker and adding 8.3 mL of 18 M H2SO4. This beaker was placed in a larger beaker containing cold water and then put on a stir plate and continuously stirred throughout the creation of the solution. 6.3g of citric acid monohydrate were added and a pH meter was placed into the solution to determine the starting pH. Through multiple trials it was determined that a pH of 4 and a running time of 1 hour produced nanotubes with desirable qualities that were well attached to the titanium samples [6]. In order to reach a final pH of 4, 11 to 12 g of NaOH pellets were first mixed into the solution which usually resulted in a pH around 2. Afterwards 0.63g of NaF was added to the solution and this was followed by the addition of more NaOH until the desired pH of 4 was reached.

When the solution had cooled to 32.5oC it was ready to be used and poured into the 250 mL beaker that contained the two electrodes. A piece of plastic was taped over the top of the beaker as a safety precaution and the beaker was placed back on the stir plate. A multimeter was placed in series with a power source so the current running through the cell could be monitored. The negative lead from the power source was placed on the platinum or the cathode while the positive lead was placed on the wire attached to the titanium sample. The power source applied a potential of 20V (DC) across the cell. This potential was applied for 1 or 2 hours depending on the desired nanotube length.

To confirm the presence of nanotubes on the surface of the titanium Field Emission Scanning Election Microscopy (FESEM) was used to image the samples. Each sample had images taken at three different sites. At each site images were collected at three magnifications including: 74,000x, 30,000x, and 12,000x. These images were then analyzed to determine the diameter and wall thickness of the nanotubes. A small piece of the titanium samples was removed with the diamond saw. This slice was then bent into a semicircular shape using two pairs. The bent slice was examined with the SEM to find the length of the nanotubes. All FESEM data and measurements are reported in the REU student report by Alyson Michael.

Cold Spray

The .74mm thick titanium samples were polished with 600 grit silicon carbide paper to create a rough surface to enhance the attachment of the cold spray particles. 40 micron titanium particles were accelerated and sprayed in a thin layer onto the surface of the titanium sample using helium. The temperature at the beginning of the spraying was ~325oC while the ending temperature was ~345oC. These processing temperatures were well below the melting temperature of titanium (1660oC) which is why it was regarded as a cold process.

Like the plain titanium samples the cold spray samples were placed in a beaker with an F- containing electrolyte to create TiO2 nanotubes. These samples were also imaged using the FESEM to confirm the presence of nanotubes on the cold spray surface. Both the pure cold spray and cold spray with nanotubes samples were characterized chemically using Energy Dispersive Spectroscopy (EDS). The surface roughness of the cold spray was also characterizeed using profilometery.

Since the cold spray was sprayed by hand there was not uniform thickness of the coating across the surface of the titanium. To examine how much the cold spray thickness varied, samples were cut in half and mounted so their cross section was visible. These samples were polished to remove saw marks and scratches. The cross sections were etched using a Kroll’s reagent that contained 1 mL HF, 5mL HNO3, and 50 mL of deionized water. After 10-20 seconds the reagent was washed off the sample with water. This etching was done so the boundary between the titanium substrate and cold spray would be clear. An optical microscope was used to image and measure the thickness of the cross sections.

Cell Culture and Staining

For cell culture 3T3 cells or mouse fibroblasts were used. Two frozen cell vials were removed from the cryobiological storage vessel and placed in a water bath that had been warmed to 37oC for thawing. All equipment including the cell vials was wiped with 70% ethanol before being placed into the biological hood. A 15 mL centrifuge tube was filled with 10 mL of complete growth media which contained 10% fetal bovine serum (FBS) and 1% penicillin-streptomycin (PS). The cells were pipetted from their vial into the centrifuge tube and then 1 mL of that suspension was placed back in the vial and pipetted up and down a few times to rinse the vial before it was moved back to the tube. This suspension was then centrifuged for 5 minutes at 1200 rpm to create a pellet. Back in the hood the centrifuge tube was inverted over a beaker containing bleach to remove the media. With only the pellet left in the tube the bottom of the centrifuge tube was flicked until the cell pellet broke apart. 5 mL of complete media was then added back to the tube and 1-2 mL was gently drawn up and down while the tip of the pipette was at the conical end of the tube. Following that the entire contents of the tube were drawn into the pipette and released a few times before being moved to a culture flask. To place the suspension into the culture flask the suspension was released from the pipette against the growing surface in a side to side motion so the cells are spread out. An additional 6-8 mL of media was then placed into the culture flask before it was sealed and placed into an incubator with a temperature of 37oC [7].

The cell culture flask was examined every other day to make sure the cells were dividing and that the sample was not contaminated. At these times the cells were also fed. Feeding included removing half of the old media from the flask and replacing it with new media that had been heated to 37oC. On Fridays extra media was added to ensure the cells had enough nutrients over the weekend. After approximately a week of feeding the cells they would achieve the desired level of confluence of 80-95% and be ready for placement onto samples [7].

Once ready the flask was placed into the biological hood and all of the media was removed with a Pasteur pipette attached to vacuum. This should be done carefully so as not to pull the cells off of the growing surface. The cells were then washed by adding 5 mL of DPBS to the flask and then rocking it back and forth so the entire growing surface is coated. This liquid was vacuumed out and washing process was repeated two more times. Then 1 mL of trypsin-EDTA was added to the culture flask and again it is rocked to ensure all cells were coated with the trypsin. The flask was moved to the incubator for a few minutes and when the cells were trypsinized enough they slide back and forth across the surface when the flask is rocked. Back under the hood, 8-10 mL of media was added to the flask and gently drawn up and down to create a cell suspension. This suspension was then moved to a centrifuge tube and DPBS was added to make the volume in the tube 15 mL. The suspension was centrifuged for 5 minutes at 1200 rpm and again when finished the media was dumped over a beaker of bleach and the pellet had to be broken and resuspended in the same process used previously [7].

With the cells again suspended they are ready to be counted. 0.5 mL of the cell suspension and 0.5 mL of trypan blue are added to a new tube and mixed together. Then 10 µL of this suspension should be added to each side of a hemocytometer. After placing the hemocytometer on a microscope the number of living cells on the four corner and center squares were counted. The number of dead or blue cells was also counted to make sure there was not a problem with the culture. This same counting process was then done for the second side of the hemocytometer. The number of living cells across the ten squares was totaled and divided by ten to find an average. To account for the dilution of the cell suspension with the trypan blue the average must be multiplied by 2 and then by 104 to find the cells per mL of the suspension. Using this number the amount of suspension to be placed on each of the samples was determined [7].

To prepare samples for cell culture they were sonicated for 5 minutes in 70% ethanol and then placed in the autoclave for a half hour at 260 oF. The samples used for cell culture included polished titanium, cold spray, titanium with nanotubes, and cold spray with nanotubes. Since the samples had an area of 1.5 cm2 they were placed individual circular dishes. The calculated amount of cell suspension was then placed on each of the samples and they were placed in the incubator for an hour to give the cells time to attach. After this hour 5 mL of media was added to each of the sample dishes. The cells were fed every other day for a week at which point the cell culture was stopped [7].

In order to determine how the 3T3 cells responded to the different surfaces they were placed on, cell staining procedures were used. The alizarin red staining procedure was used to look for calcium deposits. Samples that underwent alizarin red staining were removed from media and rinsed with DPBS. The samples were then placed in 10% buffered formalin for 10 minutes to fix the cells to the surface of the samples. After removal from the formalin the samples were repeatedly rinsed with distilled water. Enough alizarin red to coat the surface of the samples was then applied and allowed to sit for approximately 4 minutes. Excess alizarin dye was then removed from the samples. Each sample was dipped into acetone 20 times before being dipped in 50/50 acetone-xylene solution 20 times. Samples were then cleared in xylene and ready for examination with a microscope. Calcium deposits will appear a red-orange color. A second staining procedure looking for calcium salts called Von Kossa staining was also used to confirm the results of the alizarin red stain. This stain causes calcium salts to appear black or brown with the cells appearing reddish pink in color [7].

The third stain that was planned to be used was the alkaline phoshatase (ALP) stain, however due to time constraints this was not performed on samples. This stain shows the activity of ALP an enzyme which made by osteoblasts and is an indicator that bone growth is taking place. The ALP procedure used the ALP kit 86R from Sigma-Aldrich. To create a citrate-acetone-formaldehyde fixative, 25 mL citrate solution, 65 mL acetone, and 8 mL of 37% formaldehyde were mixed together. This solution was then added to 45 mL of deionized water. Then 1 mL sodium nitrate solution was added to 1 mL of FRV-alkaline solution making a diazonium salt solution. 1 mL of Naphthol AS-Bl alkaline was added to the diazonium solution to create a dye. Once the fixative solution was at room temperature the sample was immersed for 30 seconds so the cells would become fixed. The samples were then rinsed with deionized water before being immersed in the dye solution for 15 minutes. During the dyeing process the samples should not be in direct light. The samples should again be rinsed with deionized water this time for 2 minutes. They should then be counter stained with hematoxylin for 2 minutes before rinsing with tap water. After allowing the samples to air dry any ALP activity will appear purple in color [7].


Table I: This table contains data compiled from nanotube samples that were run for 1 hour with a starting pH of 4. For further explanation of the choice of these parameters and the rest of the nanotube data refer to “Processing and Characterization of Hierarchical Surface Treatments for Ti Implants” by Alyson Michael. This paper explains that nanotube length is greatly influenced by anodization time and the pH of the electrolyte solution [8,6]. These characterizations describe the nanotube only samples that were used in cell culturing.


Tube Diameter (nm)

Tube Thickness (nm)

Tube Length (nm)





Standard Deviation




Standard Deviation of the Mean




Figure 1 A: FESEM image showing TiO2 nanotubes grown on a polished titanium sample using an F- containing electrolyte solution with a potential of 20V, pH of 4, for a time of 1 hour. The tubes were pretty uniform across the surface of the sample with the dark circles representing the opening of the tube or a pore. B: Nanotubes at a lower magnification than shown in A. The whiter areas appear to be raised regions meaning polishing was not completely uniform.


c:\documents and settings\christi mclinn\desktop\reu\important for paper\nanotube pictures\31s_395_1hr_30kx_site1.tif c:\documents and settings\christi mclinn\desktop\reu\important for paper\nanotube pictures\31s_395_1hr_74kx_site2.tif


Figure 2 A: Again shows TiO2 nanotubes grown at a potential of 20V, pH of 4, for 1 hour on a sample that was later used for cell culturing. B: Shows this sample like figure 1B does not have a totally flat surface with the white patches showing differences in the surface height. It is important to note there are also some voids where there are no nantubes which is shown with the red arrows. C: This is a cross section of the nanotubes and FESEM images like this one were used to determine the average length of nanotubes for each sample. D: A zoomed out image of the way the nanotube coating broke off of the surface after the sample was bent into a semicircular shape during the cross sectioning procedure. c:\documents and settings\christi mclinn\desktop\reu\important for paper\nanotube pictures\sample 10_23kx1.jpg

c:\documents and settings\christi mclinn\desktop\reu\important for paper\nanotube pictures\sample 10_61kx1.jpg


c:\documents and settings\christi mclinn\my documents\downloads\10s_600x_surface1.tif

c:\documents and settings\christi mclinn\my documents\downloads\10s_35kx_length_21 (1).tif

Figure 3 A: This image taken with a digital camera shows the titanium sample after a thin layer of cold spray of 40 µm titanium particles has been added. The original titanium surface can be seen on the right edge. B: The plan for the laser deposition was to create a grid pattern like the one shown below. Due to time constraints cold spray was the only microstructure used.

h:\documents and settings\1822602\local settings\temporary internet files\content.word\imag0138.jpg

c:\documents and settings\christi mclinn\desktop\reu\important for paper\cold spray sample.jpg


Table II: Surface roughness of cold sprayed samples measured with a hand held profilometer (Mahr Pocket Surf PS1). Where Ra is the arithmetic mean roughness and Rz is the peak to valley height.

Cold Spray Sample 1

Cold Spray Sample 2


Ra (µm)

Rz (µm)


Ra (µm)

Rz (µm)





































Standard Deviation



Standard Deviation



Figure 4 A: This FESEM image shows a cold sprayed sample which demonstrates that the 40 µm particles are not uniformly spread across the surface, some areas have larger mounds of particles than others. B: A cold spray sample with some unidentified extra particles which appear white. These may be due to contamination of the cold spray particles which will be further discussed later. It is also a good example of the Rz value or peak to valley distance.


c:\documents and settings\christi mclinn\desktop\reu\important for paper\sem coldspray\sem\coldspray_1kx1.tif c:\documents and settings\christi mclinn\desktop\reu\important for paper\sem coldspray\sem\coldspray_132x1.tif

Figure 5 A: Is an image taken with an optical microscope of a cold spray cross section after it has been etched with a Kroll’s reagent. The area above the line is the cold spray titanium particles while the area below is the original titanium sample. B: Another cold spray cross section and at this magnification scratches from cutting the cross section can clearly be seen.


c:\documents and settings\christi mclinn\desktop\reu\important for paper\optical microscope pictures for paper\ti coldspray interface2.tif c:\documents and settings\christi mclinn\desktop\reu\important for paper\optical microscope pictures for paper\ti coldspray interface1.tif

Figure 6 A: Cold spray sample cross section with plain titanium on the bottom which should have a thickness around 740 µm. The top layer of cold spray does not have a very uniform thickness across the entire sample as demonstrated by the large differences between some of the measurements. B: A second cold spray cross section has a more uniform coating in the area being examined but the layer is thicker than seen with the first cross section sample shown in A.


c:\documents and settings\christi mclinn\desktop\reu\important for paper\optical microscope pictures for paper\ti coldspray measurements1.tif c:\documents and settings\christi mclinn\desktop\reu\important for paper\optical microscope pictures for paper\ti coldspray measurement3.tif

Figure 7: This is a line scan of a cold spray sample created using an SEM Energy Dispersive Spectrometer (EDS). The graphs show CPS (counts per second) which measures the intensity of the chemicals present on the sample over the distance of the sample shown in µm. As expected there is a high level of titanium throughout but once into the cold spray region of the line scan (left of red line, at about 700 µm) there is a considerable amount of aluminum present suggesting the titanium cold spray particles were likely contaminated.

Cold Spray

Titanium Substrate

Figure 8: A second cold spray cross section line scan created using EDS. The composition appears pretty similar to the first sample with peaks of aluminum in the cold spray region of the sample but this region begins more around 740 µm. This difference in distances is likely due to a difference in the amount of polishing each sample was subjected to. There is also a large difference in the amount of cold spray present since this sample has a much greater amount of titanium from cold spray than figure 7.

Titanium Substrate


Cold Spray

Figure 9 A: Sample 26 was first cold sprayed and then put through nanotubes processing with a potential of 20V, pH of 4, for a time of 1 hour. B: Nanotubes were present on the surface that previously contained only cold spray. Light and dark regions in the picture were due to the uneven surface the tubes grew on since there were areas where the cold spray was much thicker.


c:\documents and settings\christi mclinn\desktop\reu\important for paper\sem coldspray\sem\26cs1_700x_2.tif c:\documents and settings\christi mclinn\desktop\reu\important for paper\sem coldspray\sem\26cs1_74kx_site4.tif

Figure 10 A: Another cold spray with nanotubes sample. In the center of the picture is a valley with a smooth surfaced structure and it is on this type of structure that the nanotubes were present. 10B: A zoomed in picture of the smooth surface showing TiO2 nanotubes were successfully grown on top of a titanium cold spray.


c:\documents and settings\christi mclinn\desktop\reu\important for paper\sem coldspray\ifoundsample28\28cs2_352_2hr_6kx_site3.tif c:\documents and settings\christi mclinn\desktop\reu\important for paper\sem coldspray\ifoundsample28\28cs2_352_2hr_75kx_site3.tif

Figure 11: Electron Image 6 shows the surface of sample with both cold spray an nanaotubes as it is being analyzed using EDS. Spectrum 12 displays the elements present in the nanotube valley while spectrum 11 is the analysis of the cold spray peak. This sample appears to have less contamination than others as it doesn’t have any copper or iron. The samples are similar in that they have mostly the same elements present but the peak has a lot less titanium and oxygen and a lot more fluorine, sodium than the valley. The valley that has TiO2 nanotubes has more oxygen which is expected and only a little bit of aluminum which also makes sense since there is not much cold spray apparent in that region.



Wt% Sigma























Wt% Sigma
























Figure 12 All of these images are alizarin red stains and were taken with a microscope. A: Is a plain polished titanium sample where the pinkish red area represents calcium deposits from cells. The stain is not a very deep red meaning there was not a huge amount of cell activity on the surface. Most of the cells seem centered around the more scratched areas since cells tend to like growing in corners or crevices. B: This sample had TiO2 nanotubes and shows areas with a deep red color indicating a lot of cell activity in a few select areas. C: This sample had a cold spray surface and similar to B showed cell activity in patches and not very uniformly. The pink blotches where there were a lot of calcium deposits likely formed around cold spray peaks where there was greater roughness. D: Finally the last sample contained both nanotubes and cold spray and had by far the most uniform calcium deposits indicating the cells were probably more evenly distributed across the surface. The color of this sample was also darker than most of what was seen on the other samples meaning there was likely more calcification happening when both parts of the hierarchical structure were present.


c:\documents and settings\christi mclinn\desktop\bio\31ala3.jpg c:\documents and settings\christi mclinn\desktop\bio\tipol_ala3.jpg

c:\documents and settings\christi mclinn\desktop\bio\coldspray_ala3.jpg


c:\documents and settings\christi mclinn\desktop\bio\26cs1_ala2.jpg

Figure 13 All of these samples have been stained with the von kossa staining procedure. Calcium salts are stained brown or black while the cytoplasm is stained pink. A: This was a polished titanium sample with some calcium deposited and traces of cells but like figure 12 it was not very uniform across the surface. B: The sample contained only a nanotube coating; the image was very dark which made it very difficult to determine where there were deposits. The pink stains of the cell cytoplasm are easier to pick out for this sample. C: This was a sample with a cold spray coating and again it was quite dark so harder to analyze. There is very little pink and some unusual blue areas which were likely a reaction to something that was present in the cold spray. D: Lastly the cold spray with nanotubes sample showed some calcium deposits as well as more cell cytoplasm. Similar to figure 12D the reaction appeared more uniform across the surface than what was seen with the other samples.

c:\documents and settings\christi mclinn\desktop\bio\10vk2.jpg


c:\documents and settings\christi mclinn\desktop\bio\tipol_vk2.jpg


c:\documents and settings\christi mclinn\desktop\bio\28cs2_vk3.jpg c:\documents and settings\christi mclinn\desktop\bio\cs1_vk2.jpg


TiO2 nanotubes were grown on top of a polished titanium surface using anodic oxidation with a potential of 20V and a pH of 4. This process was allowed to proceed for 1 hour and produced nanotubes with an average length of about 671 nm, a tube thickness of 28 nm, and an average tube diameter of about 78 nm. These parameters were used to successfully create a fairly uniform layer of TiO2 nanotubes. Some SEM images like figure 2B show voids in the nanotube surface where either something prevent their growth or caused patches to fall off. At times the tubes appeared to be white in SEM images because they were on a different level from the rest of the nanotube coating. This difference in height may have been due to poor polishing of the titanium surface or it could also have been caused by too short of a running time for the oxidation. If the nanotubes are not given enough time they do not reach the sort of equilibrium length where they even out. Although it would be been ideal to use the nanotubes samples that underwent anodic oxidation for 2 hours these samples were more fragile than the 1 hour samples. For unexplained reasons these two hour samples that had nanotubes that were about 1 µm in length repeatedly fell off of the titanium surface, which is why the 1 hour samples were used instead.

To create a microstructure, helium was used to accelerate 40 µm particles of titanium so they would stick onto a titanium substrate that had been polished with 600 grit silicon carbide paper. Testing with a profilometer showed these samples have an average roughness of about 7.4 µm. Ideally this value would be 200-250 µm to create the desired cell interactions. In order to increase this roughness value a larger sized titanium particle should be tested with the cold spray. The peak to valley distance of the cold spray surface was found to be approximately 41 µm. This value makes sense as it is about the size of the individual titanium particles.

Cross sections of the cold spray samples were used to further examine the thickness and uniformity of the cold spray surface. Cold spray mounted in epoxy and polished to remove the large scratches from the surface. Samples were then etched using a Kroll’s reagent so the boundary between the original titanium and the cold spray titanium would be clear. Using an optical microscope these two regions were then examined. It was interesting to find the original titanium surface measurements to range from 750-780 µm when the sample started at 740 µm thick. I would have expected that after polishing the titanium sample should have had a thickness slightly less than 740 µm and not more. The cold spray thickness ranged anywhere from 400 to 700 µm thick. These cross sections were also examined using the SEM so that line scans could be obtained for the samples. Line scans showed that there was aluminum on the surface beginning in the 700 to 740 µm range which is more expected. The aluminum contamination signaled the beginning of the cold spray and this distance was more in line with what the sample thickness should have been. Aluminum concentration varied throughout the cold spray surface with peaks every couple hundred micrometers. There was no clear pattern to the intensity of the aluminum peaks.

Figure 9 shows that TiO2 nanotubes were successfully grown on top of a cold spray titanium surface. At this point these samples do not have nanotubes distributed uniformly across their surface. These samples had very bumpy regions that appeared to predominantly cold spray as well as more valley like regions with smoother structures present. Figure 10 displays these smoother structures and upon zooming in with the SEM it was discovered that these areas had nice nanotube coatings. It is important to note that these cold spray with nanotube samples had an odd precipitate layer form on top of their surface. At a pH of 4 it was a very thick layer and white in color. In an attempt to reduce the amount of precipitate a pH of 3.5 was used and this did greatly reduce the layer. It is difficult to say whether this precipitate layer was forming because of the cold spray itself or because the cold spray contained so many other elements like aluminum, copper, and iron.

SEM EDS was then used to examine the chemical composition of the cold spray with nanotube samples in both peak regions which appeared cold spray like and valley regions which contained nanotubes. Figure 11 shows that the valley area contained a lot more oxygen than seen in the peak areas, which is logical since the presence of TiO2 nanotubes can account for that difference. There are much larger concentrations of fluorine, sodium, and aluminum in the peak or cold spray regions. This could be due to the microstructure roughness of the cold spray trapping ions during the anodic oxidation process.

In order to determine the biological response to the hierarchical structure, plain titanium, nanotube only, cold spray only, and cold spray with nanotube samples were all used for cell culturing. Mouse fibroblasts (3T3-1s) were left on these samples for 1 week and were fed with complete media every other day. At the end of the week half the samples were stained with alizarin red and the other with von kossa. The alizarin red stain turns calcium deposits a reddish pink color. While this color was found on all samples the polished titanium sample had only a very light pink color which was only on part of the surface meaning there was less osteoblast activity on the plain titanium. The plain nanotube sample had a deeper color meaning more calcium was present but again this was only in patches and not across the surface. So the TiO2 nanotubes seemed to have more calcium than the regular titanium but only in certain places. The cold spray only sample again had patches of more calcium which was likely due to areas where there was a lot of built up cold spray and greater surface roughness. Finally the cold spray with nanotube sample shown in figure 12 D had the most pink and therefore calcium show up after the staining. The calcium deposits were not just present in circular areas but across most of the surface indicating the mouse cells were spreading more widely across the surface of this sample than they had for the others. This more uniform distribution of calcium is indicative of more mineralization by the cells which is desirable for orthopedic applications. Although there was an undesirable precipitate layer on this sample the cell alizarin red stain seems to indicate a preference a surface than contains both a micro and nanostructure. These results are fairly preliminary and further trials must be done to confirm these findings.

The von kossa staining images were less clear than those taken for the alizarin red stain. For this stain calcium deposits are stained a dark color while the cytoplasm of the cell is dyed pink. It was more difficult to analyze because of the dark color of the cold spray as well as the overall darkness of the images themselves as seen through the microscope. Figure 13 A appears to have dark patches where calcium has been deposited. Unfortunately, the images of the nanotubes and cold spray alone are very unusual looking and difficult to interpret. The sample with both cold spray and nanotubes seems to have more calcium as well as cytoplasm on the surface across the entire surface of the sample. These results seem to be following the same trend as the alizarin red stained samples did, but the von kossa data seems much more inconclusive and open to interpretation.


In order to improve oseeointegration between titanium orthopedic implants and patient’s bone, this project proposed hierarchical surface modifications. These surface modifications included a microstructure made up of 40 µm titanium particles from cold spray, as well as a nanostructure of TiO2 created using anodic oxidation. Each structure was characterized individually before they were combined for biological testing purposes. It can be concluded that it is possible to create TiO2 nanotubes on top of a titanium cold spray layer. At this point the nanotubes are not yet distributed uniformly. There was a contamination of the cold spray samples with other particles such as aluminum, copper, and iron but the preliminary results are promising. The cell culture and alizarin red staining seemed to show more calcium deposits on the sample with the hierarchical structure than any of the others indicating a larger amount of mineralization. Results from the von kossa staining were less clear but seemed to follow the same pattern seen with the alizarin red stain. In the future more extensive biological testing should be done as well as fluorescent microscopy so that the cells can be monitored throughout the culture to look for spreading and the amount of differentiation. The use of laser deposition to create a microstructure could also solve problems with cold spray contamination and lack of uniformity across the sample surface. Once adjustments have been made this testing should be repeated on titanium alloys like Ti-6Al-4V. Although these results are only preliminary, they may indicate in the future this type of hierarchical structure can be added to orthopedic implants to improve ossointegration.


  1. Variola, F., Brunski, J. B., Orsini, G., Tambasco de Oliveira, P., Wazen, R., & Nanci, A. (2011). Nanoscale surface modifications of medically relevant metals: State of the art and perspectives.Nanoscale, 2, 335-353.

  2. Tran, N., & Webster, T. J. (2009). Nanotechnology for bone materials. Wiley Interdisciplinary Reviews: Nanomedicine and Nanobiotechnology, 1(3), 336-351.

  3. Das, K., Bose, S., & Bandyopahyay, A. (2009). Tio2 nanotubes on ti: Influence of nanoscale morphology on bone cell-materials interaction. Journal of Biomedical Materials Research Part A, 90(1), 225-237.

  4. Fuerst, J. D. (2012). Laser powder deposition of ti-6al-4v, ti-15mo, ti/ta, and almgb14/tib2 for orthopedic devices. (Unpublished doctoral dissertation, South Dakota School of Mines and Technology).

  5. Oh, S., Daraio, C., Chen, L., Pisanic, T. R., Finones, R. R., & Jin, S. (2006). Significantly accelerated osteoblast cell growth on aligned tio2 nanotubes.Journal of Biomedical Materials Research, 78A, 97-103.

  6. Michael, A.  Processing and characterization of hierarchical surface treatments for ti implants. Unpublished raw data, Department of Materials and Metallurgical Engineering, South Dakota School of Mines and Technology, Rapid City, USA.

  7. Black, M. Cell culturing lab guide. Unpublished manuscript, Biomedical Engineering, South Dakota School of Mines and Technology, Rapid City, USA.

  8. Crawford, G., & Chawla, N. (2009). Porous hierarchical tio2 nanostructures: Processing and microstructure relationships. Acta Materialia, 57, 854-867.


We acknowledge financial support from the National Science Foundation (REU Site Award #1157074) Thank you to Dr. Sears and the cell lab especially Dani Black for her help with cell culturing. Also a special thanks to Dr. Grant Crawford, Dr. Michael West, and Dr. Alfred Boysen for their guidance throughout the program. Finally thank you to Alyson Michael and Ellen Sauter for their collaboration on this project.

Directory: reu
reu -> Connecting learning objects to instructional design theory: a definition, a metaphor, and a taxonomy
reu -> 8000 Series Respirator Product Copy – Australia/New Zealand 8000 Series Facepiece and Accessories Your Face Never Had It So Comfortable
reu -> Template Field Operations Guide
reu -> Construction of a Tunable External Cavity Diode Laser and a Far Off-Resonant Optical Dipole Trap for Ultracold Atoms Joan Dreiling nsf – reu 2006 Advisor: Carl Wieman Scott Papp
reu -> Frequency stabilization of a diode laser by means of temperature stabilization and passive resonant optical feedback
reu -> Functional Model of a Human Middle Ear nsf-uc 2012-2013 Academic-Year reu program
reu -> Cancer Research Project
reu -> Rheumatic Fever etiology
reu -> Danish Penal Law in the Middle Ages Cases of Homicide and Woundings
reu -> Issues in Argument Structure Department of Linguistics

Share with your friends:

The database is protected by copyright © 2019
send message

    Main page