Abstract
Bone damage patients may suffer from metal toxicity resulting from an adverse reaction. To avoid the need for a second operation, we set out to identify a material that can be used as a substitute for metal in small fragment plates, which is compatible with the human body. Thus, in this study, we set out to study the development of a material that can be applied to the small fragment plate, based on a hydroxyapatite (HA)-polylactic acid (PLA) composite. This study examined three main factors, namely, the ratio of the PLA to the HA, injection temperature (T) and injection pressure (P). Based on the ASTM standard, the best results (Code 4 and Code 1) obtained from the mechanical property tests (tension and flexural) were 44.02 MPa and 63.97 MPa, respectively. When compared to HA-HDPE, our material offers both strength and biodegradable/biocompatible advantages. By inspection with scanning electron microscope (SEM) and energy-dispersive spectrometry (EDS), we could identify the components of the HA and distribution pattern. In terms of biocompatibility, Code 1 is promising. To maximize the composite desirability, optimal condition was mathematically calculated. In addition, finite element analysis confirmed that the proposed bone fixation plate would not be damaged when the wrist is impacted.
1 Introduction
Damage to the bones as a result of an accident is relatively common, regardless of gender or age and this issue is gradually becoming more acute. Data collected from the Chiang Mai-Ram Hospital between 2010 and 2012 regarding the treatment of bone fractures in various parts of the body revealed that the most common types of fracture are those of the elbow and forearm, followed by those of the wrist and the hand, then the shoulder and upper arm [1]. Generally, bone fractures will spontaneously reconnect. Treatment involves pulling the bones into position and then binding them until they have completely reconnected. This may take 1–3 months depending on the age of the patient, and the position and nature of the fracture(s). There are many types of bone fracture treatment, depending on the position of the fractures and their nature. In the treatment of bone cracks or fractures, surgery involves the use of a bone fragment plate. Once the patient’s own bone has regenerated, secondary surgery is needed to remove the artificial bone plate, which is usually metallic and, therefore, cannot decompose in the human body. This repetitive surgery can lead to infection, and even metal distortion and resulting pain if the patients spend time in an air-conditioned room or otherwise cool place. Even though the metal plate should be harmless, if inserted in some locations, it may migrate to another location. This can give rise to issues called by metal corrosion, which can lead to the deterioration of the thin coating on the plate.
Several materials have been developed as possible substitutes for metal. The objective of medical materials is to be used as a substitute for lost body parts. At the same time, the method should be safe, trustworthy, economical and harmless. Ceramics are usually adopted for this purpose. Various types of ceramics are widely used as substitutes and reinforcements for bone, especially hydroxyapatite (HA); HA and materials based on calcium phosphate, which have mineral components that are similar to those of human bone and which are, thus, more compatible with the human body than other types of ceramic. Research into economical artificial nano-materials has shown that porous HA has a strong texture and, therefore, can be used in the human body for repairing damaged bones. On the other hand, mechanical limitations of artificial bone-based HA materials will be an issue when comparing with metallic or ceramic based. As a consequence, biocompatibility and mechanical properties of artificial bone-based HA/HA composites led to our interest.
The processing, characterization and mechanical properties of HA high-density polyethylene (HA-HDPE) were reported by Wang et al. [2]. They found that an increasing HA in a polymer matrix will lead to an increase in Young’s modulus and tensile strength; however, it will reduce the strain to fracture and a ductility-brittle transition when HA reaches 30 volume%. Bakar et al. [3] proposed a composite material-based polymer. Bioactive composites comprising synthetic HA particulate and semi-crystalline polyetheretherketone (PEEK) polymer were investigated. The tensile strength and Young’s modulus of HA-PEEK composites ranged within the bony tissue. These results mean that HA-PEEK is promising as an alternative for load-bearing orthopedic application. Although a wide range of polymer matrices have been promoted, polylactic acid (PLA) and its composites have been identified as offering considerable potential for such applications. PLA is a plastic made from fermented plant material such as cornstarch, which offers a promising alternative to petroleum-based plastics. The absorption of carbon by plants is one possible means of minimizing the greenhouse effect that leads to global warming. Furthermore, PLA does not generate toxic gas when burned. The synthesis of PLA was initially developed by researchers at the DuPont Company of the USA. PLA is crystal clear and glossy as a result of additives in the formulation. It also offers mechanical properties and can be used as a basic polymer with thermoplastic properties. PLA can retain odor and taste. Furthermore, it is resistant to oil and fat, while oxygen, carbon dioxide, and water can pass through it easily. It has a low impact strength, which is similar to that of polyvinyl chloride (PVC) without a plastic enhancer. Its strength, impact strength and flexibility are similar to those of polyethylene terephthalate (PET). In addition, PLA has properties that are similar to those of polystyrene (PS) and can be adjusted such that its properties are close to those of polyethylene (PE) or polypropylene (PP). Therefore, the basic properties of PLA can be adjusted such that its basic forming and usage properties are equal to those of polyolefin, which is made using a petrochemical process. PLA is a polymer with various attractive properties and which can be applied to a wide range of value-added plastics in many areas such as medicine. PLA is a biodegradable polymer. It is compatible with the tissue of the human body and is ultimately absorbed by the body. This makes PLA an excellent choice for medical applications. It has long been used for medical tools, such as sutures, staples, wound dressings, surgical implants, orthopedic fixation devices and drug conveyers or releasers, which can effectively control the rate and duration of drug release. The properties of PLA are listed in Table 1.
Properties of PLA.
| Property | Nature Work® PLAa | Biomer® L9000b |
|---|---|---|
| Density (g/ml) | 1.24 | 1.25 |
| Tg (°C) | 56.7–57.9 | n/a |
| Tm (°C) | 140–152 | n/a |
| HDT (°C) | 40–45 (amorphous) 135 (crystalline) | n/a |
| Tensile strength (MPa) | 53 | 70 |
| Elongation (%) | 6 | 2.4 |
| Flexural modulus (MPa) | 350–450 | 3600 |
aDatasheet: NatureWork PLA, NatureWork LLC, USA. bDatasheets: Biomer L9000, Biomer, Netherlands.
In spite of PLA exhibiting excellent mechanical properties, the range of its application is limited because it is difficult to control the hydrolysis rate and limit its poor hydrophilicity. To overcome these problems, blends of PLA with other materials have been studied, e.g. starch [4] and poly(ε-caprolactone) (PCL) [5]. HA is an inorganic part of naturally occurring bone. In the human body, minerals or inorganic compounds constitute around 69% of the bone weight, with HA being the main component. Thus, it is commonly used as a bone substitute. HA can be prepared from any of three sources, as follows: (1) from chemicals such as calcium nitrate and ammonium phosphate, (2) from coral, which is processed to transform it into HA. As coral is a scarce resource, and the process incurs environmental problems, this process is not commonly adopted and (3) from animal bones such as those of cows, buffalos, elephants, fish, pigs and also human beings. The bones of cows and buffalos are often used as they are plentiful, given that humans exploit these species for motive power and as a food source. Generally, HA can be used in any of three ways: powder form, rigid form and porous form.
This study addressed the rigid form of HA, the design and development of the small fragment plates that are commonly used to treat accidents and bone diseases. The study examined the use of computer programs to assist with the design and testing needed to derive an appropriate and safe product. This included the identification of a suitable formulation for the HA and PLA, and the undertaking of a biological reaction test.
2 Materials and methods
2.1 Composited material
PLA:HA, which is a composite material was studied. The mechanical properties of a composite material such as elastic modulus, Poisson’s ratio, mass density, tensile strength and yield strength are needed from the experiment. To find the composite material’s mechanical properties, PLA and HA were mixed well and loaded into a twin screw horizontal injection machine (Battenfeld type BA 250 CDC 40 tons screw diameter 22 mm) (Wittmann, Vienna, Austria) at the required temperature (T) and pressure (P), with an injection speed of 370 rpm, an injection holding P of 3 bar, and an injection holding time for each piece of 45 s. The tensile test was performed using dumbbell-shaped parts as specified in ASTM D638 type I in the testing machine, a Baldwin Universal Testing System (Instron, PA, USA) (10 tons), at a T of 25°C, an extraction rate of 5 mm/min, and a gripping distance of 109 mm [6]. The bending test was conducted using bar-shaped parts conforming to ASTM D790, using a WYKEHAM-FARRANCE (Controls Group, Milan, Italy) testing machine (10 tons). The experiment was conducted for three areas at a T of 25°C, a movement rate of 2 mm/min, and a base distance of 80 mm [7].
2.2 Design of experiments
To conduct the experiments, we identified the experimental factors as being the PLA, HA, T and injection P. The PLA:HA ratios can be divided into three ratios with percentages of 90:10, 85:15 and 80:20 by weight. The factors for the injection molding process are listed in Table 2.
High-level and low-level parameters.
| No. | Parameter | High level (+1) | Low level (−1) | Unit |
|---|---|---|---|---|
| 1 | T | 185 | 170 | °C |
| 2 | P | 60 | 55 | Bar |
The correlation of the experimental factors was then analyzed: information was extracted using a graph plotter program, trends in the data were monitored using regression analysis, and the appropriate points of the experimental factors were determined from the results of the mechanical and biocompatibility tests.
2.3 Electron microscope analysis and energy-dispersive X-ray spectrometer (EDS)
A study was made on the microstructure of the surface of the part: samples were attached to a stub and then coated with gold in a coating machine using a low-vacuum scanning electron microscope (SEM) (JEOL JSM-5910LV) to analyze the elemental composition through the application with the energy-dispersive spectrometry (EDS) method.
2.4 Cytotoxicity test
The biocompatibility test was done by studying the cytotoxicity of the samples. Cartilage cells (SW 1353) were put together with the samples in a test tube for 24 h. Then, the survival rate of the cells in each tube was determined from the light absorption of the cells using a spectrophotometer at wavelengths of 540 nm and 630 nm, from which the cell viability percentage was calculated according to equation (1). A cell proliferation test is used to study the reaction of a cell toward a chemical under test. Nowadays, there are many means of implementing cell proliferation tests, such as counting the number of living cells, measuring the degree of increase in the DNA, and checking the metabolic activity, the latter being the easiest and most popular method.
The metabolic activity measurement can determine whether a type of cell is capable of increasing in quantity. Those cells that exhibit considerable metabolic activity will be capable of exhibiting an increase in the number of living cells. There are many assays for metabolism measurement, e.g. MTT, MTS and XTT cell proliferation assay (colorimetric assay). The principle or method of measurement involves measuring the reduction environment (mitochondrial reductase) of the mitochondria in cells. MTT, a yellow tetrazole, is reduced to purple formazan in living cells [8]. XTT has been proposed to replace MTT, yielding higher sensitivity and a higher dynamic range. The formed formazan dye is water soluble, avoiding a final solubilization step. The MTS assay is often described as a “one-step” MTT assay, which offers the convenience of adding the reagent straight to the cell culture without the intermittent steps required in the MTT assay. However, this convenience makes the MTS assay susceptible to colorimetric interference as the intermittent steps in the MTT assay remove traces of colored compounds, while these remain in the microtiter plate in the one-step MTS assay. Precautions are needed to ensure accuracy when using this assay, and there are strong arguments for confirming the MTS results using qualitative observations under a microscope [9]. The absorption rate is examined at 570 nm. An increase in the intensity of the purple corresponds to an increase in the number of living cells. However, the limit on the use of MTT is the measurement of the number of cells at the end point. This is because, after the MTT is reduced to formazan, it must be dissolved in dimethyl sulfoxide (DMSO), which prevents the cells used in the MTT process from being reused.
2.5 Finite element analysis
The properties of a HA-PLA composite with an elastic modulus, Poisson’s ratio, mass density, tensile strength and yield strength were manipulated from the experimental data and used in a FEM simulation model. Next, we performed our simulation by Solidworks 2013 (Dassault Systemes Solidworks Corporation, MA, USA) finite element method (FEM) module: a model of a T-shaped plate with a screw diameter of 3.5 mm was constructed with a CAD/CAM program conforming to ISO 9001:2008, EN ISO 13485:2012, T-plate 3.5, right angled (three head holes) stainless steel 141.103. A drawing of the design of a bone fixation plate is presented in Figure 1, assuming the use of a 3.5-mm screw. We determined the direction of the force on the sample, assuming a fall with the hands outstretched onto the ground. Finally, we determined the forces on the sample resulting from the external load option to examine the force on the plate with loads of 600 N, 300 N and 150 N based on a correlation of a human falling and an area of impact as shown in Table 3.
Bone fixation plate with a 3.5-mm screw.
Correlation rate of fall and impact on the forearm, elbow and wrist [10].
| Male | Female | |
|---|---|---|
| Distance of fall (m) | 3.3528 | 3.3528 |
| Velocity on impact (m/s) | 8.114 | 8.114 |
| Deceleration distance (m) | 0.3048 | 0.3048 |
| Rate of deceleration (m/s2) | 107.899 | 107.899 |
| Weight (N) | 765.09 | 609.41 |
| Force of impact (N) | 8.416 | 6.7302 |
| Time (s) | 0.8266 | 0.8266 |
| Area of impact (m2) | 8.06×10−5 | 8.06×10−5 |
| P of impact (N/m2) | 1.044×105 | 0.835×105 |
3 Experimental results
3.1 Mechanical properties
Based on a full factorial design with three mixtures, two injection Ts and two injection Ps gives rise to 12 experiments. The injection molding process of a HA-PLA composite conforming to ASTM D638 Type I and ASTM D790 was examined using dumbbell- and bar-shaped molds, respectively. The size of the dumbbell was 165-mm long and 3-mm thick, while the bar was 13-mm wide, 127-mm long and 3-mm thick. We implemented 12 experiments with 10 replications of the injection molding. The average and standard deviations of the mechanical properties, tensile strength and flexural strength were determined to be as listed in Table 4.
Average and standard deviations of tensile and bending strengths.
| Std. order (Code) | PLA:HA (weight%) | T (°C) | P (bar) | Ultimate tensile | Ultimate flexural strength | ||
|---|---|---|---|---|---|---|---|
| x̅ (MPa) | SD | x̅ (MPa) | SD | ||||
| 1 | 90:10 | 170 | 55 | 42.88 | 1.42 | 63.97 | 0.49 |
| 2 | 90:10 | 170 | 60 | 39.52 | 0.32 | 60.58 | 0.34 |
| 3 | 90:10 | 185 | 55 | 38.72 | 3.04 | 59.76 | 1.73 |
| 4 | 90:10 | 185 | 60 | 44.02 | 1.75 | 61.89 | 1.41 |
| 5 | 85:15 | 170 | 55 | 35.76 | 2.52 | 58.75 | 0.83 |
| 6 | 85:15 | 170 | 60 | 34.52 | 0.66 | 59.06 | 0.73 |
| 7 | 85:15 | 185 | 55 | 31.34 | 1.41 | 59.45 | 0.05 |
| 8 | 85:15 | 185 | 60 | 30.15 | 1.94 | 57.44 | 2.18 |
| 9 | 80:20 | 170 | 55 | 29.44 | 0.74 | 56.67 | 1.24 |
| 10 | 80:20 | 170 | 60 | 31.98 | 0.83 | 55.62 | 1.20 |
| 11 | 80:20 | 185 | 55 | 28.79 | 3.11 | 56.34 | 1.23 |
| 12 | 80:20 | 185 | 60 | 27.87 | 0.32 | 55.54 | 0.99 |
From Table 4, we can determine that the workpiece of the fourth experiment with a PLA:HA ratio of 90:10% by weight, an injection T of 185°C, and an injection P of 60 bar gives the mean of the highest tensile strength of 44.02 MPa and that of the 11th experiment with a PLA:HA ratio of 80:20% by weight, an injection T of 185°C, and an injection P of 60 bar, which gives the mean of the lowest tensile strength at 27.87 MPa. The tensile strength of HA-PEEK (semi-crystalline polyetherketone) as reported by Bakar et al. [3] and HA-HDPE (high-density polyethylene) as reported by Wang et al. [2] ranged from 49.0 MPa to 83.3 MPa and 17.3–20.7 MPa, respectively. Our composite material HA-PLA, at 27.9–44.0 MPa, falls between the two. Compared to previous composite materials, the tensile strength of our proposed material is moderate and matches the lower limits of cortical bone. Although our HA-PLA material has a lower tensile strength than HA-PEEK, it is biodegradable and biocompatible, unlike HA-PEEK. Compared to HA-HDPE, our material offers both strength and biodegradable/biocompatible advantages. From the results of the experiment, a stress–strain curve could be constructed as shown in Figure 2.
Stress-strain curve of HA-PLA composite.
From Figure 2, we can see the correlation between the stress and strain in the tensile test, which can be used to determine the characteristics of the polymer workpiece by comparing the characteristics of the stress–strain curve of the different polymer groups. From the results of the bending test shown in Table 4, we can see that the workpiece used in the first experiment with a PLA:HA component ratio of 90:10% by weight, an injection T of 170°C, and an injection P of 55 bar can best withstand the bending force before incurring damage. The average of the ultimate flexural strength is 63.97 MPa. The maximum bending distance before breaking is 0.015 m. Furthermore, the workpiece with the lowest flexural strength is that used in the seventh experiment in which the PLA:HA component ratio is 80:20% by weight, the injection T is 185°C, and the injection P is 60 bar, giving the lowest average flexural strength of 55.54 MPa.
3.2 Electron microscope analysis
Images of the workpiece surface taken by a SEM at 15 kV, a magnification of 500×, and a scale of 50 μm, as well as elemental analysis using an energy-dispersive X-ray spectrometer (EDS), identified the morphology of the HA-PLA composite produced by the injection molding process, as shown in Figures 3 and 4 .
Workpiece surface and elemental measurement (Code 4).
Workpiece surface and elemental measurement (Code 1).
Inspection of the work surface by SEM shows that there were some spots on the surface of every workpiece, especially the Code 4 and Code 1 (Figures 3 and 4) workpieces that best results in tension and flexural strength, could be clearly observed at a magnification of 500×. It could be observe that HA particles were well dispersed, with a homogeneous distribution in the polymer matrix. An elemental analysis of the pellets identified the components of the HA, specifically the phosphate (P) and calcium (Ca), as described in Table 5. The molecular formula of the HA is Ca10(PO4)6(OH)2. Theoretically [11], the Ca/P ratio by weight and by atomic are 2.151 and 1.67, respectively, for pure HA. From Table 5, we can report that the Ca/P ratio by weight of our workpiece range from 1.47 to 2.37, and the Ca/P ratio by atomic range from 1.09 to 2.20. Workpiece Code 4, which reported the highest ultimate tensile strength, gets the Ca/P ratio by atomic near the HA in nature; however, it gets the Ca/P ratio by weight less than in nature. Code 1, which has the highest ultimate flexural strength, has a higher Ca/P ratio.
Elemental analysis by EDS.
| Element | O | P | Ca | Ca/P ratio |
|---|---|---|---|---|
| Code 1 | ||||
| Weight% | 27.56 | 2.65 | 6.27 | 2.37 |
| Atomic% | 24.24 | 1.04 | 2.29 | 2.20 |
| Code 2 | ||||
| Weight% | 28.47 | 3.68 | 7.45 | 2.02 |
| Atomic% | 25.50 | 2.89 | 5.62 | 1.94 |
| Code 3 | ||||
| Weight% | 28.54 | 2.76 | 6.25 | 2.26 |
| Atomic% | 25.49 | 1.80 | 3.89 | 2.14 |
| Code 4 | ||||
| Weight% | 28.92 | 4.56 | 9.25 | 2.03 |
| Atomic% | 26.28 | 2.12 | 3.59 | 1.69 |
| Code 5 | ||||
| Weight% | 32.68 | 7.65 | 17.45 | 2.28 |
| Atomic% | 30.98 | 6.56 | 13.88 | 2.12 |
| Code 6 | ||||
| Weight% | 35.59 | 5.94 | 14.28 | 2.40 |
| Atomic% | 34.95 | 3.01 | 5.60 | 1.86 |
| Code 7 | ||||
| Weight% | 31.90 | 19.90 | 29.17 | 1.47 |
| Atomic% | 41.90 | 13.47 | 17.18 | 1.26 |
| Code 8 | ||||
| Weight% | 49.34 | 13.78 | 21.90 | 1.59 |
| Atomic% | 62.78 | 9.80 | 12.07 | 1.23 |
| Code 9 | ||||
| Weight% | 50.33 | 16.08 | 23.63 | 1.47 |
| Atomic% | 65.66 | 12.40 | 13.59 | 1.09 |
| Code 11 | ||||
| Weight% | 49.22 | 15.59 | 24.09 | 1.55 |
| Atomic% | 67.45 | 10.87 | 12.95 | 1.19 |
| Code 12 | ||||
| Weight% | 51.45 | 15.78 | 24.55 | 1.56 |
| Atomic% | 69.56 | 11.45 | 13.78 | 1.20 |
3.3 Biocompatibility as determined by cytotoxicity analysis
The cytotoxicity test used cartilage cells (SW 1353) in the HA-PLA composite and involved measuring the cell viability percentage relative to a control by measuring the metabolism of cells (MTT assay). We investigated 12 conditions completed with 10 replications. The %Survival rate of the cells is shown in Table 6.
%Survival rate of the cells as measured by cytotoxicity test.
| Code | %Survival rate of the cells | SD |
|---|---|---|
| Code 1 | 143 | 1.69 |
| Code 2 | 137 | 1.94 |
| Code 3 | 139 | 0.01 |
| Code 4 | 136 | 1.32 |
| Code 5 | 115 | 3.75 |
| Code 6 | 111 | 0.98 |
| Code 7 | 106 | 0.78 |
| Code 8 | 108 | 2.93 |
| Code 9 | 110 | 0.68 |
| Code 10 | 108 | 1.29 |
| Code 11 | 100 | 3.36 |
| Code 12 | 107 | 1.35 |
The %Survival rate of the cells >100% means that the number of survival cells in that particular sample is greater than the control, or the particular sample performs better than the control. From Table 6, we can see that the HA-PLA composite with a component ratio of 90:10% by weight, at a T of 170°C, and an injection P of 55 bar (Code 1) produced the highest %Survival rate of the cells (equation 1); however, with moderate standard deviation. A comparison of the results of the %Survival rate of cell and tensile tests reveals that the workpiece with the best tensile strength and that with the best %Survival rate of cell are not the same. The Code 4 workpiece has the best tensile strength at 44.02 MPa, while the Code 1 workpiece has the best %Survival rate of cell. Henceforth, workpieces with the best tensile strengths will be selected for the analysis of the bone fixation plate.
4 Analysis of statistical experiments
Data from the experiments were used to analyze the appropriation of the experimental factors in order to determine the correlation between the independent variables (ratio of PLA to HA powder, injection T and injection P) and the dependent variables (tensile test, bending test and percentage cell viability). A computer program for performing a regression analysis enabled the appropriation of the regression types, as follows.
4.1 Normal distribution analysis
The data residue was used to plot a graph of the normal probability and determine the data distribution. This gives the result shown in Figure 5; the normal probability plot of the residue of the tensile strength, the flexural strength and the cell viability percentage. All three graphs show that the data are distributed in a straight line. There were no data distributed far away from the line. Thus, we can conclude that the residual distribution is in the normal distribution form.
Normal probability plot of residues.
4.2 Variance stability analysis
An analysis was made of the residual distribution plot, as shown in Figure 6. The residues of the tensile strength, flexural strength and percentage cell viability reveal that the data were distributed in both a positive and negative way regardless of the response value. This points to the variance stability of the data.
Residual distribution.
4.3 Independence assumption analysis
A scatter plot analysis is shown in Figure 7. The residues and experimental order of the tensile strength, flexural strength and percentage cell viability show that the data were randomly scattered, with no specific tendency. This means that the data is independent and not dependent on the sequence of the experiments.
Residues and experimental order.
From the three analysis methods mentioned above, we found that the residual data was normally distributed. The variance was stable and independent. Then, the formulated regression models based on this information were found to be sufficient to represent the dependent variables.
4.4 Regression model analysis
A regression analysis was performed using the Minitab (Minitab Inc., PA, USA) program to examine the correlation between the independent and dependent variables. The results are listed in Tables 7–9 . A variance analysis was conducted to confirm the efficiency of the predicted regression model. The results are listed in Tables 10–12 .
Coefficient analysis of tensile strength.
| Term | Coefficients | SE coefficients | T | P |
|---|---|---|---|---|
| PLA | 72.7 | 14.05 | * | * |
| HA | 643.9 | 475.62 | * | * |
| PLA*HA | −984.0 | 659.43 | −1.49 | 0.232 |
| PLA*T | 14.5 | 14.05 | 1.03 | 0.377 |
| HA*T | 462.4 | 475.62 | 0.97 | 0.403 |
| PLA*HA*T | −658.0 | 659.43 | −1.00 | 0.392 |
| PLA*P | 9.0 | 14.05 | 0.64 | 0.568 |
| HA*P | 302.9 | 475.62 | 0.64 | 0.570 |
| PLA*HA*P | −421.0 | 659.43 | −0.64 | 0.569 |
| R2 | 93.66% | R2 (adj) | 76.74% |
*Could not be calculated but significant.
Coefficient analysis of flexural strength.
| Term | Coefficients | SE coefficients | T | P |
|---|---|---|---|---|
| PLA | 68.03 | 9.031 | * | * |
| HA | 46.90 | 305.641 | * | * |
| PLA*HA | −48.50 | 423.765 | −0.11 | 0.916 |
| PLA*T | −2.82 | 9.031 | −0.31 | 0.775 |
| HA*T | −48.04 | 305.641 | −0.16 | 0.885 |
| PLA*HA*T | 73.50 | 423.765 | 0.17 | 0.873 |
| PLA*P | 0.12 | 9.031 | 0.01 | 0.990 |
| HA*P | 8.80 | 305.641 | 0.03 | 0.979 |
| PLA*HA*P | −14.50 | 423.765 | −0.03 | 0.975 |
| R2 | 87.87% | R2 (adj) | 55.52% |
*Could not be calculated but significant.
Coefficient analysis of cell viability percentage.
| Term | Coefficients | SE coefficients | T | P |
|---|---|---|---|---|
| PLA | 270 | 15.96 | * | * |
| HA | 3408 | 540.11 | * | * |
| PLA*HA | −4944 | 748.86 | −6.60 | 0.007 |
| PLA*T | 6 | 15.96 | 0.38 | 0.729 |
| HA*T | 185 | 540.11 | 0.34 | 0.754 |
| PLA*HA*T | −280 | 748.86 | −0.37 | 0.733 |
| PLA*P | −5 | 15.96 | −0.29 | 0.789 |
| HA*P | 64 | 540.11 | 0.12 | 0.914 |
| PLA*HA*P | −49 | 748.86 | −0.07 | 0.952 |
| R2 | 98.93% | R2 (adj) | 96.08% | |
*Could not be calculated but significant.
Variance analysis of tensile strength.
| Source | DF | Seq SS | Adj SS | Adj MS | F | P |
|---|---|---|---|---|---|---|
| Regression | 8 | 321.051 | 321.0506 | 40.1313 | 5.54 | 0.093 |
| Linear | 1 | 276.830 | 11.0768 | 11.0768 | 1.53 | 0.304 |
| Quadratic | 1 | 16.138 | 16.1376 | 16.1376 | 2.23 | 0.232 |
| PLA*HA | 1 | 16.138 | 16.1376 | 16.1376 | 2.23 | 0.232 |
| Component*T | ||||||
| Linear | 2 | 17.793 | 7.8499 | 3.9249 | 0.54 | 0.630 |
| PLA*T | 1 | 13.857 | 7.7369 | 7.7369 | 1.07 | 0.377 |
| HA*T | 1 | 3.936 | 6.8494 | 6.8494 | 0.95 | 0.403 |
| Quadratic | 1 | 7.216 | 7.2161 | 7.2161 | 1.00 | 0.392 |
| PLA*HA*T | 1 | 7.216 | 7.2161 | 7.2161 | 1.00 | 0.392 |
| Component*P | ||||||
| Linear | 2 | 0.119 | 2.9963 | 1.4982 | 0.21 | 0.824 |
| PLA*P | 1 | 0.110 | 2.9626 | 2.9626 | 0.41 | 0.568 |
| HA*P | 1 | 0.009 | 2.9392 | 2.9392 | 0.41 | 0.570 |
| Quadratic | 1 | 2.954 | 2.9540 | 2.9540 | 0.41 | 0.569 |
| PLA*HA*P | 1 | 2.954 | 2.9540 | 2.9540 | 0.41 | 0.569 |
| Residual error | 3 | 21.742 | 21.7424 | 7.2475 | ||
| Total | 11 | 342.793 |
Variance analysis of flexural strength.
| Source | DF | Seq SS | Adj SS | Adj MS | F | P |
|---|---|---|---|---|---|---|
| Regression | 8 | 65.0355 | 65.03547 | 8.12943 | 2.72 | 0.222 |
| Linear | 1 | 60.6651 | 0.01515 | 0.01515 | 0.01 | 0.948 |
| Quadratic | 1 | 0.0392 | 0.03920 | 0.03920 | 0.01 | 0.916 |
| PLA*HA | 1 | 0.0392 | 0.03920 | 0.03920 | 0.01 | 0.916 |
| Component*T | ||||||
| Linear | 2 | 2.2661 | 1.37003 | 0.68501 | 0.23 | 0.808 |
| PLA*T | 1 | 1.5925 | 0.29131 | 0.29131 | 0.10 | 0.775 |
| HA*T | 1 | 0.6736 | 0.07395 | 0.07395 | 0.02 | 0.885 |
| Quadratic | 1 | 0.0900 | 0.09004 | 0.09004 | 0.03 | 0.873 |
| PLA*HA*T | 1 | 0.0900 | 0.09004 | 0.09004 | 0.03 | 0.873 |
| Component*P | ||||||
| Linear | 2 | 1.9715 | 0.01299 | 0.00649 | 0.00 | 0.998 |
| PLA*P | 1 | 1.8959 | 0.00055 | 0.00055 | 0.00 | 0.990 |
| HA*P | 1 | 0.0756 | 0.00248 | 0.00248 | 0.00 | 0.979 |
| Quadratic | 1 | 0.0035 | 0.00350 | 0.00350 | 0.00 | 0.975 |
| PLA*HA*P | 1 | 0.0035 | 0.00350 | 0.00350 | 0.00 | 0.975 |
| Residual error | 3 | 8.9788 | 8.97883 | 2.99294 | ||
| Total | 11 | 74.0143 |
Variance analysis of cell viability percentage.
| Source | DF | Seq SS | Adj SS | Adj MS | F | P |
|---|---|---|---|---|---|---|
| Regression | 8 | 2591.47 | 2591.470 | 323.934 | 34.66 | 0.007 |
| Linear | 1 | 2087.23 | 334.312 | 334.312 | 35.77 | 0.009 |
| Quadratic | 1 | 407.39 | 407.386 | 407.386 | 43.59 | 0.007 |
| PLA*HA | 1 | 407.39 | 407.386 | 407.386 | 43.59 | 0.007 |
| Component*T | ||||||
| Linear | 2 | 68.24 | 1.467 | 0.734 | 0.08 | 0.926 |
| PLA*T | 1 | 60.63 | 1.355 | 1.355 | 0.15 | 0.729 |
| HA*T | 1 | 7.62 | 1.099 | 1.099 | 0.12 | 0.754 |
| Quadratic | 1 | 1.31 | 1.307 | 1.307 | 0.14 | 0.733 |
| PLA*HA*T | 1 | 1.31 | 1.307 | 1.307 | 0.14 | 0.733 |
| Component*P | ||||||
| Linear | 2 | 27.27 | 26.522 | 13.261 | 1.42 | 0.368 |
| PLA*P | 1 | 4.98 | 0.801 | 0.801 | 0.09 | 0.789 |
| HA*P | 1 | 22.04 | 0.130 | 0.130 | 0.01 | 0.914 |
| Quadratic | 1 | 0.04 | 0.040 | 0.040 | 0.00 | 0.952 |
| PLA*HA*P | 1 | 0.04 | 0.040 | 0.040 | 0.00 | 0.952 |
| Residual error | 3 | 28.04 | 28.039 | 9.346 | ||
| Total | 11 | 2619.51 |
In Tables 7–12, regression models corresponding to each dependent variable could be formulated based on a confidence level of 95%, as follows:
Ultimate tensile strength=72.7PLA+643.9HA−984PLA*HA+14.5PLA*T+462.4HA*T−658PLA*HA*T+9PLA*P+302.9HA*P−421PLA*HA*P (2)
Ultimate flexural strength=68.03PLA+46.90HA −48.50PLA*HA–2.82PLA*T−48.04HA*T+73.50PLA *HA*T+0.12PLA*P+8.80HA*P−14.50PLA*HA*P (3)
% viability=270PLA+3408HA−4944PLA*HA+6PLA *T+185HA*T−280PLA*HA*T−5PLA*P+64HA*P− 49PLA*HA*P (4)
4.5 Response optimization analysis
When manufacturing a bone fixation plate, all the dependent variables must be simultaneously considered, thus, requiring a multi-objective approach. A response-optimizer module, which considers this approach, is used to maximize the composite desirability. We regarded all the independent variables as being equally important, with all the targets set to the maximum value obtained from the experiment. The results are listed in Table 13.
Response optimization of HA-PLA composite.
| Parameters | Goal | Lower | Target | Upper | Weight | Import |
|---|---|---|---|---|---|---|
| Ave. ultimate tensile strength | Maximum | 27.87 | 44.02 | 44.02 | 1 | 1 |
| Ave. ultimate flexural strength | Maximum | 57.46 | 63.97 | 63.97 | 1 | 1 |
| % Viability | Maximum | 93.86 | 142.55 | 142.55 | 1 | 1 |
| Global solution | ||||||
| Components | ||||||
| PLA=90 | ||||||
| HA=10 | ||||||
| Process variable | ||||||
| Temp=170 | ||||||
| Press=55 | ||||||
| Predicted responses | ||||||
| Ave. ultimate tensile strength=40.715 | Desirability=0.795356 | |||||
| Ave. ultimate flexural strength=62.590 | Desirability=0.836299 | |||||
| % Viability=142.093 | Desirability=0.989172 | |||||
| Composite desirability=0.869758 | ||||||
The identification of the optimal value of the independent variables using the response-optimizer function revealed that, upon adding more HA, the tensile strength and flexural strength were reduced. This caused the desirability of the response to fall accordingly. When PLA was added, however, the tensile strength and flexural strength were found to have increased. Furthermore, the desirability of the response optimization also increased. The appropriate ratio of the components for the injection molding of HA-PLA composite was 10% HA by weight, 90% PLA by weight, an injection T of 170°C, and an injection P of 55 bar with the predicted response of the tensile strength, flexural strength and cell viability percentage of 40.715 MPa, 62.590 MPa and 142.09%, as well as a desirability of 0.795, 0.836 and 0.989, respectively. The composite desirability was 0.869758, as shown in Figure 8.
Factor optimization.
5 Finite element analysis of bone fixation plate
From the objectives of the research into the development of a bone fixation plate made of a HA-PLA composite, the data on the mechanical properties calculated from the experimental results shown in Table 14 were used to analyze the loading strength of the bone fixation plate based on the size of a standard fixation plate used in bone trauma patients.
HA-PLA composite properties.
| Properties | Value | Unit |
|---|---|---|
| Elastic modulus in X | 440.20 | MPa |
| Poisson’s ratio in XY | 0.0609 | – |
| Shear modulus in XY | – | N/m2 |
| Mass density | 1411.63 | kg/m3 |
| Tensile strength in X | 44.02 | MPa |
| Compressive strength in X | – | N/m2 |
| Yield strength | 38.87 | MPa |
| Thermal expansion coefficient in X | – | /K |
| Thermal conductivity in X | – | W/(m·K) |
| Specific heat | – | J/(kg·K) |
| Material damping ratio | – | N/A |
5.1 Material properties
After material testing, weighting, and volume calculation, we can obtain the properties of the HA-PLA composite used to analyze the bone fixation plate, as shown in Table 14.
An analysis of a bone fixation plate based on the properties of the HA-PLA composite was conducted using a simulation software. We can clearly see the failure resulting from the tensile force applied to the area, as described in Table 15.
Tensile force of bone fixation plate.
| Force (N) | Defection | von Mises (MPa) | Yield strength (MPa) | FOS |
|---|---|---|---|---|
| 600 |  | 127.61 | 38.87 | 0.30 |
| 300 |  | 66.17 | 38.87 | 0.59 |
| 150 |  | 33.09 | 38.87 | 1.2 |
Table 15 indicates that the von Mises stress appeared in the plate when subjected to a force of 150 N, equal to 33.09 MPa. This arose around the screw area. A plate manufactured using the proposed composite material had a yield strength of 38.87 MPa, as determined from the results of the experiments. Thus, a force of 150 N produced a stress that did not cause any damage to the plate. The factor of safety (FOS) of the plate was determined and was found to be 1.2. For 600 N and 300 N, the results of the simulation reported that the von Mises stress exceeds the yield strength of the material, meaning that the plate would be damaged if this amount of force was to be applied.
The value of the correlation of the fall and its impact, as listed in Table 3, shows that if a man falls with a force of 765 N, he will incur a stress around his wrist of approximately 8.42 N or 0.1044-MPa P of impact, which is comparable to the stress value of the fixation plate made from a HA-PLA composite, which has a strength of 33.09 MPa when subjected to a force of 150 N. This means that the proposed bone fixation plate will not be damaged when the wrist is impacted in the event of a fall. In addition to the mechanical properties of the HA-PLA composite bone fixation plate, another advantage of the plate relative to a conventional metal bone fixation plate is its biocompatibility, as confirmed by a cytotoxicity analysis.
6 Conclusions
This research set out to study the mechanical properties and biological reaction of a HA-PLA composite. The HA phase of the injection-composited plate was determined by a morphological examination. The mechanical properties, as determined by experiment, were used in a finite element analysis of the bone fixation plate. A plate made from a HA-PLA composite exhibited damage toward the top of the plate with von Mises stress in the screw region. However, simulation results showed that this composite plate could withstand a force of 765 N caused by a person falling and incurring stress around his/her wrist. The composite plate is also non-toxic to the human body, as determined by a cytotoxicity analysis. This composite material will allow the development of bone fixation plates that can take the place of metal versions.
Article note:
This paper should not be disseminated without the written permission of the authors.
Acknowledgments
Chiang Mai University (CMU), through the research administration office, provided the funding.
Compliance with Ethical Standards: This study was funded by Chiang Mai University. We, the authors, were under no P or competing interests that affected our professional judgment in any way. The results of the research are ours alone, without conflict of interest. This article does not contain any studies with animals and human participants.
References
[1] Chiang Mai Ram Hospital, “Patient Data Classification by Department”, Chiang Mai, Thailand, 2012.Search in Google Scholar
[2] Wang M, Porter D, Bonfield W. Br. Ceram. Trans. 1994, 93, 91–95.Search in Google Scholar
[3] Bakar MSA, Cheang P, Khor KA. Compos. Sci. Technol. 2003, 63, 421–425.10.1016/S0266-3538(02)00230-0Search in Google Scholar
[4] Zhang JF, Sun X. Biomacromolecules 2004, 5, 1446–1451.10.1021/bm0400022Search in Google Scholar
[5] Kuo SW, Huang CF, Tung YC, Chang FC. J. Appl. Polym. Sci. 2006, 100, 1146–1161.10.1002/app.23227Search in Google Scholar
[6] ASTM D638-10, Standard Test Method for Tensile Properties of Plastics, ASTM International, West Conshohocken, PA, 2010.Search in Google Scholar
[7] ASTM D790-02, Standard Test Methods for Flexural Properties of Unreinforced and Reinforced Plastics and Electrical Insulating Materials, ASTM International, West Conshohocken, PA, 2002.Search in Google Scholar
[8] Mosmann T. J. Immunol. Methods 1983, 65, 55–63.10.1016/0022-1759(83)90303-4Search in Google Scholar
[9] Cory AH, Owen TC, Barltrop JA, Cory JG. Cancer Commun. 1991, 3, 207–212.10.3727/095535491820873191Search in Google Scholar PubMed
[10] Gary SN, Timothy DS. Physics Calculations Related to Falling Objects, Nelson & Associates Safety Engineering: Texas, 2010.Search in Google Scholar
[11] Murugan R, Ramakrishna S. Compos. Sci. Technol. 2005, 65, 2385–2406.10.1016/j.compscitech.2005.07.022Search in Google Scholar
©2018 Walter de Gruyter GmbH, Berlin/Boston
This article is distributed under the terms of the Creative Commons Attribution Non-Commercial License, which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.
