Materials Science & Engineering C 123 (2021) 111967

Contents lists available at ScienceDirect

Materials Science & Engineering C

journal homepage: www.elsevier.com/locate/msec

Simultaneous engineering of nanofillers and patterned surface macropores

of graphene/hydroxyapatite/polyetheretherketone ternary composites for
potential bone implants
Zhihuan Huang a, Yizao Wan a, b, Xiangbo Zhu a, Peibiao Zhang c, Zhiwei Yang a, Fanglian Yao d,
Honglin Luo a, b, *
a
Jiangxi Key Laboratory of Nanobiomaterials, Institute of Advanced Materials, East China Jiaotong University, Nanchang 330013, China
b
School of Materials Science and Engineering, Tianjin University, Tianjin 300072, China
c
Changchun Institute of Applied Chemistry, Chinese Academy of Sciences, Changchun 130022, China
d
Key Laboratory of Systems Bioengineering of Ministry of Education, School of Chemical Engineering and Technology, Tianjin University, Tianjin 300072, China

A R T I C L E I N F O A B S T R A C T

Keywords: Incorporating bioactive nanofillers and creating porous surfaces are two common strategies used to improve the
Graphene oxide tissue integration of polyetheretherketone (PEEK) material. However, few studies have reported the combined
Hydroxyapatite use of both strategies to modify PEEK. Herein, for the first time, dual nanoparticles of graphene oxide (GO) and
Polyetheretherketone
hydroxyapatite (HAp) were incorporated into PEEK matrix to obtain ternary composites that were laser
Porous surface
Bone implant
machined to create macropores with diameters ranging from 200 μm to 600 μm on the surfaces. The surface
morphology and chemistry, mechanical properties, and cellular responses of the composites were investigated.
The results show that micropatterned pores with a depth of 50 μm were created on the surfaces of the composites,
which do not significantly affect the mechanical properties of the resultant composites. More importantly, the
incorporation of GO and HAp significantly improves the cell adhesion and proliferation on the surface of PEEK.
Compared to the smooth surface composite, the composites with macroporous surface exhibit markedly
enhanced cell viability. The combined use of nanofillers and surface macropores may be a promising way of
improving tissue integration of PEEK for bone implants.

1. Introduction conducted and several methods have been developed to improve the
bioactivity of PEEK in dental and orthopedic applications. One of the
Implant products for total knee replacement, total hip replacement, methods is the incorporation of bioactive components such as hy­
and osteosynthesis are mainly made of metallic materials such as tita­ droxyapatite (HAp) [5], tantalum [6], and niobium pentoxide [7]. The
nium and its alloys, stainless steel, and cobalt-based alloys due to their second method is surface modification through grafting or coating
ease of storage and high mechanical strength. However, some disad­ bioactive molecules on PEEK [8]. Creating porous surface structure is
vantages such as metal ion release and stress shielding threaten their another effective strategy [9,10], because the porous surface allows for
clinical applications [1]. In this context, polymer-based composites have appropriate biological fixation, initial and long-term stability by
become the alternative due to their good biocompatibility and tunable improving bone ingrowth [11–13].
mechanical properties. Among various polymeric matrices, poly­ In our previous work, together with immobilization of dual growth
etheretherketone (PEEK) has emerged as a promising matrix material factors (insulin-like growth factor-1 and bone morphogenetic proteins-
because it shows excellent biocompatibility, outstanding mechanical 2), pores on PEEK were created by immersion in concentrated sulfuric
performances, superior biological stability, radiolucency, minimal im­ acid [8]. However, the resultant pore size (<1 μm) is too small to allow
mune toxicity, and X-radiolucency [2,3]. However, PEEK is bioinert for ingrowth of bone cells. Recently, Cai et al. created pores on the
without osteoconductivity or the ability to integrate with host bone [4], surface of PEEK/magnesium calcium silicate composite with femto­
thus limiting its use in clinical applications. Much work has been second laser, achieving a pore size of 20 μm [14]. However, previous

* Corresponding author at: Jiangxi Key Laboratory of Nanobiomaterials, Institute of Advanced Materials, East China Jiaotong University, Nanchang 330013, China.
E-mail address: hlluotju@126.com (H. Luo).

https://doi.org/10.1016/j.msec.2021.111967
Received 22 October 2020; Received in revised form 7 February 2021; Accepted 8 February 2021
Available online 12 February 2021
0928-4931/© 2021 Elsevier B.V. All rights reserved.
Z. Huang et al. Materials Science & Engineering C 123 (2021) 111967

studies indicated that pore sizes need to be larger than 150 μm to be ethanol and drying, macropores were created on the surface of np-HAp/
suitable for bone formation [12,15]. Therefore, larger macropores need GO/PEEK composite using ultra-short pulse laser etching technique (PF-
to be created on the surface of PEEK composites to improve their YLP-30 W, Pufei Technology Co., Ltd., Nanchang, China), as reported in
bioactivity. In addition, interfacial bonding between nanofillers and our previous study [20]. Three laser spot sizes of 200 μm, 400 μm, and
PEEK is also critical. A previous study demonstrated that graphene oxide 600 μm were selected to produce macropores with different sizes and the
(GO) could serve as an interface to increase the interfacial bonding resultant samples were named np-HAp/GO/PEEK-p1, np-HAp/GO/
strength between HAp and PEEK [16]. However, there is no report found PEEK-p2, and np-HAp/GO/PEEK-p3, respectively. The composite
on integrating these strategies to simultaneously add GO and bioactive without macropores was labeled as np-HAp/GO/PEEK-b.
HAp, and create porous surface such that a surface-porous PEEK com­
posite implant with improved mechanical strength and enhanced 2.5. Characterization
bioactivity can be achieved.
Herein, we report a novel ternary PEEK composite reinforced with The surface morphology was analyzed with a scanning electron mi­
both GO and lamellar HAp (LHAp) that was exfoliated to nanoplate HAp croscope (SEM, FEI Nano 430) together with an energy dispersive
(termed np-HAp) during composite preparation. Furthermore, a hier­ spectrometer (EDS), a laser scanning confocal microscope (LSCM TCS
archically porous surface was created with laser machining method. The SP8, Leica Microsystems CMS GmbH, Wetzlar, Germany), and a metal­
objectives of this work were to prepare the surface-porous ternary lographic microscope (XHC-SG1, Beijing Dongfang Huace Science and
composites, characterize the morphology, structure, and mechanical Technology Center, Beijing, China). Transmission electron microscope
properties, and to preliminarily investigate how cells respond to com­ (TEM, FEI Tecnai G2 F2) was used to examine the morphology of np-
bined modifications of LHAp and GO incorporations and creation of HAp and GO nanosheets. The morphology of GO nanosheets was
surface macropores. further characterized with an Agilent 5500 atomic force microscope
(AFM, USA) operated in taping mode. The crystalline structure was
2. Materials and methods analyzed with an X-ray diffractometer (XRD, Rigaku D/Max 2500 v/pc,
Japan) at scanning rates of 4 min− 1 (wide-angle diffraction, 10–60◦ ) and
2.1. Preparation of GO and LHAp 1 min− 1 (small-angle diffraction, 2–10◦ ). To determine the surface
chemistry, Fourier transform infrared (FTIR, PerkinElmer Spectrum
The GO aqueous dispersion was prepared by the modified Hummer’s one) analysis was performed using the ATR (attenuated total reflec­
method following the reported procedures [7]. The AFM images of the tance) mode over a spectral range of 4000–400 cm− 1 and a resolution of
obtained fully exfoliated GO nanosheets are shown in Fig. S1A and B 4 cm− 1. Water contact angles of PEEK and its composites were measured
that show typical wrinkles (Fig. S1C) with a typical height of ~0.68 nm using a liquid/solid interface analyzer (Chengde Chenghui, JGM-360B).
(Fig. S1D). The preparation procedures of LHAp were described in Tensile properties of PEEK and composites were tested using a micro
previous studies [17–19]. The SEM image of the obtained LHAp nano­ electromagnetic fatigue testing machine (MUF-1050, Tianjin Care
particles is presented in Fig. S1E and a HRTEM image of LHAp is pre­ Measure & Control Co., Ltd., Tianjin, China) and the measurement was
sented in Fig. S1F. The particle size distributions of LHAp and GO are performed at room temperature in accordance with national standard of
given in Fig. S1G and H, respectively. China (GB/T 1040.2–2006) using the dumbbell samples.

2.2. Preparation of np-HAp/GO assembly 2.6. Release of Ca and P ions

In a typical synthesis of np-HAp/GO, 6 mL of GO dispersion (2 mg To quantitatively determine the release of Ca and P ions from np-
mL− 1) was mixed with 50 mL of LHAp dispersion (24 g mL− 1) under HAp/GO/PEEK-b and np-HAp/GO/PEEK-p2, the samples (4 mm in
stirring to react for 5 h. Afterwards, the obtained suspension was kept for thickness and 10 mm in diameter) were immersed in 5 mL of deionized
12 h followed by centrifugation at 4500 rpm and washed with deionized water and incubated at 37 ◦ C under shaking at 120 rpm for 1, 3, and 7
water three times. Finally, the obtained product was freeze-dried for 48 days. An inductively coupled plasma atomic emission spectroscopy
h, yielding a light brown flocculent powder consisting of GO and np- (VARIAN 720-ES, USA) was used to quantify the concentration of Ca and
HAp. P ions in the soaking solutions.

2.3. Preparation of np-HAp/GO/PEEK composites 2.7. In vitro cell studies

PEEK powder was dried in a vacuum oven at 150 ◦ C for 24 h. 0.6 g of The MC3T3-E1 cells (mouse embryo osteoblasts) used in this work
np-HAp/GO complex and 6.0 g of PEEK were thoroughly mixed for 4 h at were purchased from Cell Bank of Chinese Academy of Sciences. Prior to
480 rpm using a ball mill followed by sieving with an 80-mesh screen. the experiments, PEEK and its composites were soaked in 75% alcohol
Subsequently, a co-rotating twin-screw extruder (screw diameter: 18 solution for 24 h, sterilized under UV radiation for 12 h, and washed
mm; L/D ¼40) attached with an injection molding machine (HAAKE, with phosphate buffer solution (PBS, Hyclone Company, USA) several
MiniLab, Thermo Scientific, Germany) was used to fabricate the np- times. The MC3T3-E1 cells were cultured on a petri dish for 3 days in
HAp/GO/PEEK composite. The extrusion was performed at a barrel MEM (minimum essential medium, Hyclone, USA) at 37 ◦ C. 1 mL of
temperature of 400 ◦ C and a screw speed of 90 rpm and the injection 0.25% trypsin was added to the petri dish to digest the cells. After 2 min
molding was operated at a melting temperature of 400 ◦ C, an injection of digestion, 5 mL of cell culture medium was added to terminate the
pressure of 600 bar, a holding pressure of 550 bar, and a holding time of digestion. The digested cells and culture medium were centrifuged at a
10 s. The resultant dumbbell samples had a total length of 75 mm, a thin speed of 1000 rpm for 5 min. The supernatant was discarded, and fresh
neck of 4 mm in length, and 2 mm in thickness, and the cylindrical culture medium was added to form cell suspension. 1 mL of cell sus­
samples were 4 mm in height and 10 mm in diameter. Pure PEEK pension was inoculated into a 48-well plate with samples and placed in a
sample, without np-HAp and GO, was prepared using the same method humidified incubator with 5% CO2 at 37 ◦ C for 1, 4, and 7 days. The
as the control material. medium was refreshed every 2 days.
After culture with cells, the samples were removed from petri dishes
2.4. Surface machining with laser treatment and washed with PBS twice. Cell membrane on the samples was then
permeated with PBS containing 0.3% Triton X-100, and stained with
After ultrasound cleaning in deionized water followed by absolute rhodamine-labeled phalloidin (CA1610, Solarbio Technology Co., LTD.,

2
Z. Huang et al. Materials Science & Engineering C 123 (2021) 111967

China) and 4′ 6-diamidino-2-phenylindole (DAPI, Shanghai Aladdin the peaks at 2θ = 18.7◦ , 20.6◦ , 22.8◦ , and 29.2◦ are due to diffractions of
Reagent Co., LTD., China) for 30 min. The stained cells were observed planes (110), (111), (200), and (211) of PEEK, respectively [25,26].
under an inverted fluorescence microscope (TS-2, Nikon, Japan). Although HAp peaks are obvious, the peak of GO is very weak due to its
The cell morphology was further examined by SEM. After incubation low content (1 wt%) in the np-HAp/GO/PEEK composites [16].
under the aforementioned conditions for 24 h, MC3T3-E1 cells were The FTIR results of various materials are presented in Fig. 3A and B.
fixed with 4% paraformaldehyde solution at 4 ◦ C for 12 h, dehydrated in In the spectrum of GO (Fig. 3A), peaks at 3436 cm− 1, 1739 cm− 1, 1631
a graded series of ethanol (30, 40, 50, 60, 70, 80, 90, and 100%), freeze cm− 1, and 1050 cm− 1 are due to O–H stretching, C– – O stretching vi­
dried, sputter-coated with gold, and observed with SEM. bration, C– – C stretching, and C–O–C stretching vibration, respectively
Cell proliferation was assessed using Cell Counting Kit-8 (CCK-8) [27]. There are five peaks in the spectrum of LHAp. The peak at 3439
assay after cell seeding for 1, 4, and 7 days. 200 μL of 10% CCK-8 cm− 1 is the –OH stretching vibrational band [28]; peaks at 2923 cm− 1
(Beijing Solebo Technology Co., Ltd., China) was added to each well and 2853 cm− 1 come from C–H vibration [29]; peaks at 1091 cm− 1 and
under dark conditions. After incubation for 2 h, the liquid was quickly 962 cm− 1 are due to P–O stretching vibration [30]. Compared to GO
transferred to a 96-well culture plate. The absorbance (OD) was deter­ and LHAp, the –OH band in the spectrum of np-HAp/GO shows a red
mined at 450 nm wavelength with a microplate reader (iMark, Bio Rad shift to 3428 cm− 1 and an increased intensity of 2923 cm− 1 and 2853
Co., Ltd., Hercules, USA). cm− 1 peaks, indicating hydrogen bonding between np-HAp and GO
[31,32]. In addition, there is electrostatic interactions between negative
groups in GO and Ca2+ in HAp [33]. In the spectrum of PEEK (Fig. 3B),
2.8. Statistical analysis
the vibration of diphenylketone is noted at 1734 cm− 1 and 1651 cm− 1,
and the vibration of PEEK skeleton ring at 1595 cm− 1 [34]. In addition,
All experiments were repeated at least five times to collect data,
the bands at 1220 cm− 1, 1184 cm− 1, and 1157 cm− 1 correspond to the
which was analyzed with SPSS Statistics. Results were considered sta­
vibration band of C–O–C [35], and the bands at 765 cm− 1, 840 cm− 1,
tistically significant when probability (p) values are less than 0.05.
and 863 cm− 1 are attributed to the C–H vibration of aromatic plastics
[35]. Compared to bare PEEK, the carbonyl C– – O band in np-HAp/GO/
3. Results and discussion
PEEK shifts from 1651 cm− 1 to 1645 cm− 1, indicating that the –OH
groups of np-HAp/GO may form chemical bond with C– – O of PEEK
3.1. Preparation and structure characterization
[34,36]. The possible reaction mechanism is schematically illustrated in
Fig. 3C. It is worth noting that the surface-porous np-HAp/GO/PEEK-p2
Fig. 1A illustrates the typical preparation procedures of np-HAp/GO/
composite exhibits obvious peak fluctuation at approximately 3650
PEEK-p composites starting with the fabrication of LHAp and GO.
cm− 1, which may be due to the increased exposure of –OH groups on the
Notably, all processes including extrusion, injection molding, and laser
surface [37].
machining are scalable, which is promising in future commercialization.
Unlike bare PEEK that is light, np-HAp/GO/PEEK-p and np-HAp/GO/
PEEK-b composites are dark (Fig. 1B). 3.2. Surface morphology of np-HAp/GO/PEEK composites
Fig. 2 presents the XRD results of LHAp, GO, EK, np-HAp/GO/PEEK-
b, and np-HAp/GO/PEEK-p composites. The XRD pattern from small- Fig. 4 shows the surface SEM images of PEEK, np-HAp/GO/PEEK-b,
angle diffraction (Fig. 2A) shows three characteristic peaks at diffrac­ and np-HAp/GO/PEEK-p composites. The surfaces of bare PEEK and np-
tion angles of 2.84◦ , 5.66◦ , and 8.49◦ , confirming the highly ordered HAp/GO/PEEK-b composite are relatively smooth (Fig. 4A and B). On
structure of LHAp, which agrees with our previous results [17,21–23]. the surfaces of np-HAp/GO/PEEK-p composites, many macropores of
The wide-angle result (Fig. 2B) is also consistent with our previous re­ roughly uniform size, which are arranged in an array, are observed. The
sults, which indicates the crystalline structure of np-HAp [17]. Fig. 2C average macropore sizes measured by Nano Measurer 1.2 software are
shows the XRD pattern of bare GO, which has a prominent peak at 10.5◦ , approximately 192 μm, 396 μm, and 574 μm for np-HAp/GO/PEEK-p1,
corresponding to (001) plane and a d spacing of 0.8 nm [24]. In Fig. 2D, np-HAp/GO/PEEK-p2, and np-HAp/GO/PEEK-p3, respectively. As

Fig. 1. (A) Schematic showing the preparation of np-HAp/GO/PEEK-p composites. (B) Digital photos of PEEK, np-HAp/GO/PEEK-b, and np-HAp/GO/PEEK-p.

3
Z. Huang et al. Materials Science & Engineering C 123 (2021) 111967

Fig. 2. (A) Small-angle (0–10◦ ) and (B) wide-angle (10–60◦ ) XRD patterns of LHAp. (C) XRD pattern of GO. (D) XRD patterns of bare PEEK, np-HAp/GO/PEEK-b, and
np-HAp/GO/PEEK-p composites.

shown in the insets of Fig. 4C–E, there are numerous small pores in each 3.3. Mechanical properties of np-HAp/GO/PEEK composites
macropore. The average pore sizes are 20.7 ± 8.7 μm (Fig. S2A), 36.1 ±
12.0 (Fig. S2B), and 37.8 ± 14.0 μm (Fig. S2C) for np-HAp/GO/PEEK- Fig. 5A reveals that all materials show similar stress-strain curves, a
p1, np-HAp/GO/PEEK-p2, and np-HAp/GO/PEEK-p3, respectively. roughly linear straight line followed by obvious yielding. As expected,
Similar pore structure was reported by Cai et al. [14]. The EDS results adding 10 wt% np-HAp and 1 wt% GO increases the tensile strength by
shown in Fig. 4F confirm the uniform distribution of np-HAp nanoplates approximately 16%. Previously, Ma et al. claimed that the tensile
throughout the np-HAp/GO/PEEK-p3 composite. The pore morphology strength of composites decreased by 17% when 10 wt% HAp was added
was further confirmed by optical microscope (Fig. S3) and the pore into PEEK, and it continued to decrease with further increase of HAp
depth was examined using LSCM (Fig. 4G). The pore depth is around 50 content [44]. A similar result was reported by Feng et al. in HAp/poly(L-
μm, which is larger than those (30 μm [14] and 20 μm [38]) reported by lactic acid) composite scaffolds [45]. We believe that the strong inter­
other researchers. The roughness of np-HAp/GO/PEEK-p3 was charac­ face interactions between nanofillers and PEEK, as revealed by FTIR
terized by AFM and the results reveal a roughness of 7.486 μm (Fig. S4A) result, the high strength of GO, and the uniform dispersion of nanofillers
while the roughness of np-HAp/GO/PEEK-b is only 0.198 μm (Fig. S4B). contribute to the strength improvement in the np-HAp/GO/PEEK com­
It is believed that appropriate pore depth is beneficial to cell adhesion posites. The tensile strength of np-HAp/GO/PEEK-p composites is
and nutrient transport [39] and hierarchical porous structure inside the approximately 120 MPa which does not show statistically significant
pores favors cell migration, movement, and re-migration into the space difference (p > 0.05) when compared to np-HAp/GO/PEEK-b (Fig. 5B),
between the tissue and the implant [40]. suggesting that the influence of surface laser treatment is negligible. The
The water contact angles of various materials are shown in Fig. S5. tensile modulus and elongation at break of all composites are not
PEEK shows a contact angle of 97.3◦ , indicating its hydrophobicity significantly different from bare PEEK (p > 0.05). As shown in Fig. 5B,
[4,41]. After incorporation of GO and np-HAp, np-HAp/GO/PEEK-b the tensile strength of np-HAp/GO/PEEK composites is comparable to
composite shows decreased hydrophobicity because the surfaces of GO the strength of cortical bones (100– 150 MPa [46,47]), which is
and HAp contain hydrophilic –OH groups. The np-HAp/GO/PEEK-p important from a biomechanical point of view.
composites are more hydrophilic because macropores created on the
surface lead to more exposure of hydrophilic GO and HAp. Hydrophilic
3.4. In vitro cell studies
surfaces are believed to promote cell adhesion and proliferation by
improving protein adsorption [42]. The porous surface structure may
MC3T3-E1 cells were cultured on five materials to evaluate the ef­
also contribute to the increased hydrophilicity since wettability depends
fects of GO and np-HAp incorporation and porous surface structure on
on both surface chemistry and topography [43].
cell adhesion and proliferation. Phalloidin/DAPI staining results of cells
cultured on the surfaces of PEEK and np-HAp/GO/PEEK composites
reveal fewer cells on PEEK (Fig. 6A) than np-HAp/GO/PEEK-b (Fig. 6B)
and np-HAp/GO/PEEK-p composites (Fig. 6C–E) due to the lack of

4
Z. Huang et al. Materials Science & Engineering C 123 (2021) 111967

Fig. 3. (A) FTIR spectra of LHAp, GO, np-HAp/GO. (B) FTIR spectra of PEEK, np-HAp/GO/PEEK, and np-HAp/GO/PEEK-p2 composites. (C) Schematic illustration
showing interfacial interactions between np-HAp, GO, and PEEK.

Fig. 4. SEM images of bare PEEK (A), np-HAp/GO/PEEK-b (B), np-HAp/GO/PEEK-p1 (C), np-HAp/GO/PEEK-p2 (D), and np-HAp/GO/PEEK-p3 (E). The insets show
the corresponding enlarged views. (F) Elemental mapping of Ca and P on np-HAp/GO/PEEK-p3 composite. (G) Laser scanning confocal microscopic (LSCM) image of
np-HAp/GO/PEEK-p3 composite surface.

bioactive and biocompatible np-HAp in PEEK [48]. It is believed that number of np-HAp nanoparticles exposed to surrounding medium, as
bioactive ions (such as calcium, phosphorus, and silicon) released from illustrated in Fig. 6F. The increased exposure of np-HAp nanoparticles
bioactive materials (such as HAp and bioglass) can stimulate MC3T3-E1 may lead to enhanced bioactivity of np-HAp/GO/PEEK-p composites
cells into osteoblasts [49,50]. Therefore, the increased cell number on over np-HAp/GO/PEEK-b composite. Such mechanism was also re­
the composites over bare PEEK can be attributed to the release of ported by Shuai et al. in a very recent paper [51]. Although surface
bioactive ions, which can be supported by the ion release results shown morphology has important impact on cell adhesion, migration, prolif­
in Fig. S6. Moreover, the higher cell density on np-HAp/GO/PEEK-p as eration, osteogenic differentiation, and bone regeneration [52], it is
compared to np-HAp/GO/PEEK-b is due to more released ions from the hard to identify the difference among the three laser-machined com­
np-HAp/GO/PEEK-p composites with macropores that increase the posites at this point.

5
Z. Huang et al. Materials Science & Engineering C 123 (2021) 111967

Fig. 5. Tensile properties of PEEK, np-HAp/GO/PEEK-b, and np-HAp/GO/PEEK-p. (A) Stress-strain curves. (B) Tensile strength, Young’s modulus, and elongation
at break.

Fig. 6. (A–E) Florescent images of MC3T3-E1 cells after culture on PEEK (A), np-HAp/GO/PEEK-b (B), np-HAp/GO/PEEK-p1 (C), np-HAp/GO/PEEK-p2 (D), and np-
HAp/GO/PEEK-p3 (E) for 3 days. Rhodamine-labeled phalloidin dye stained the cytoskeleton red, DAPI dye stained the nucleus blue. Scale bar: 100 μm. (F)
Illustration showing the role of macropores in improving cell proliferation on surface-porous np-HAp/GO/PEEK-p composites over np-HAp/GO/PEEK-b without
surface pores. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

Besides florescent microscopy, cell morphology was also observed cell viability on various materials. At each time point, the cell viability
with SEM (Fig. S7). After 24 h of culture, MC3T3-E1 cells spread on the on PEEK composites is significantly higher than that on bare PEEK,
surfaces of each composite. However, there is difference in cell suggesting the role of HAp and GO incorporation in enhancing cell
morphology between np-HAp/GO/PEEK-b (Fig. S7A) and np-HAp/GO/ proliferation. On day 4 and 7, np-HAp/GO/PEEK-p (except for np-HAp/
PEEK-p (Fig. S7B–D) composites. Unlike np-HAp/GO/PEEK-b, cells on GO/PEEK-p1 on day 7) exhibits significantly higher cell viability as
np-HAp/GO/PEEK-p composites show more obvious pseudopodia, compared to np-HAp/GO/PEEK-b, confirming the positive effect of
spreading around pore areas and the largest number of pseudopodia is porous surface on cell proliferation. Notably, on day 7, the cell viability
noted in np-HAp/GO/PEEK-p2. on np-HAp/GO/PEEK-p2 is significantly higher than that on np-HAp/
Fig. 7 presents the CCK-8 results that can quantitatively compare the GO/PEEK-p3. The highest cell viability on np-HAp/GO/PEEK-p2

6
Z. Huang et al. Materials Science & Engineering C 123 (2021) 111967

200 μm or 600 μm. The np-HAp/GO/PEEK-p2 composite with bioactive

nanofillers and surface macropores exhibits great potential as bone
implants.

CRediT authorship contribution statement

Zhihuan Huang: Investigation.

Yizao Wan: Writing - Original Draft, Supervision, Funding
acquisition.
Xiangbo Zhu: Visualization, Investigation.
Peibiao Zhang: Conceptualization, Writing - Proofreading.
Zhiwei Yang: Data analysis, Investigation, Formal analysis.
Fanglian Yao: Methodology, Project administration.
Honglin Luo: Writing - Review & Editing, Funding acquisition,
Supervision.

Declaration of competing interest

The authors confirm that there are no known conflicts of interest

Fig. 7. (A) Proliferation of MC3T3-E1 cells on the surface of PEEK, np-HAp/
associated with this publication and there has been no financial support
GO/PEEK-b, np-HAp/GO/PEEK-p1, np-HAp/GO/PEEK-p2 and np-HAp/GO/
PEEK-p3 composites (*p < 0.05; **p < 0.01; n = 9 in each group). for this work that could have influenced its outcome.

Acknowledgments
suggests that the macropore size of 396 μm is more suitable for cell
proliferation than 192 μm and 574 μm.
This work was supported by the National Natural Science Foundation
The influence of pore size on cellular response has been extensively
of China (Grant nos. 51662009, 31660264, and 31870963), the Key
studied [53]. However, discrepancy exists regarding the optimal pore
Research and Development Program of Jiangxi Province (No.
size, which depends on the cell and tissue types [54,55]. For instance,
20192ACB80008), and the Key Project of Natural Science Foundation of
Zhao et al. found that the 3D porous polycaprolactone (PCL) scaffolds
Jiangxi Province (20202ACBL204013).
with a pore size of 200–300 μm exhibited better cell growth, while the
scaffolds with pore size of 300–450 μm showed the best osteogenic and
Appendix A. Supplementary data
chondrogenic differentiation for human mesenchymal stem cells
(hMSCs) [56]. Kim and co-workers claimed that a pore size of 300–320
Supplementary data to this article can be found online at https://doi.
μm exhibited the best osteogenic differentiation for adipose-derived
org/10.1016/j.msec.2021.111967.
stem cells (ASCs) [57]. In another study, Oh et al. developed a pore
size gradient (90–400 μm) in fibrous PCL scaffolds and found that the
References
scaffold with a pore size range of 370–400 μm provided a more favorable
environment for chondrogenic differentiation than other pore size [1] I. Gotman, Characteristics of metals used in implants, J. Endourol. 11 (1997)
groups [58], which is consistent with our findings. Nevertheless, the 383–389.
exact underlying mechanism is not fully understood and further inves­ [2] D.F. Williams, A. McNamara, R.M. Turner, Potential of polyetheretherketone
(PEEK) and carbon-fibre-reinforced PEEK in medical applications, J. Mater. Sci.
tigation is required. Lett. 6 (1987) 188–190.
[3] S.M. Kurtz, J.N. Devine, PEEK biomaterials in trauma, orthopedic, and spinal
4. Conclusions implants, Biomaterials 28 (2007) 4845–4869.
[4] X. Xu, Y. Li, L. Wang, Y. Li, J. Pan, X. Fu, Z. Luo, Y. Sui, S. Zhang, L. Wang, Y. Ni,
L. Zhang, S. Wei, Triple-functional polyetheretherketone surface with enhanced
Novel surface-porous ternary composites consisting of GO, HAp, and bacteriostasis and anti-inflammatory and osseointegrative properties for implant
PEEK were fabricated via scalable extrusion and injection molding fol­ application, Biomaterials 212 (2019) 98–114.
lowed by laser machining. The macropores with diameters ranging from [5] R. Ma, Q. Li, L. Wang, X. Zhang, L. Fang, Z. Luo, B. Xue, L. Ma, Mechanical
properties and in vivo study of modified-hydroxyapatite/polyetheretherketone
192 μm to 574 μm and a depth of 50 μm are arranged in an array on the biocomposites, Mat. Sci. Eng. C: Mater. 73 (2017) 429–439.
surfaces of np-HAp/GO/PEEK-p composites. FTIR revealed the interfa­ [6] X. Hu, S. Mei, F. Wang, J. Qian, D. Xie, J. Zhao, L. Yang, Z. Wu, J. Wei, Implantable
cial interactions between GO, HAp, and PEEK, which endow the np- PEKK/tantalum microparticles composite with improved surface performances for
regulating cell behaviors, promoting bone formation and osseointegration,
HAp/GO/PEEK-p composites with favorable mechanical properties. Bioactive Mater 6 (2021) 928–940.
The change of the tensile strength of np-HAp/GO/PEEK-p composites is [7] N. Sharma, M. Arif, S. Monga, M. Shkir, Y.K. Mishra, A. Singh, Investigation of
not statistically significant (p > 0.05) compared to the np-HAp/GO/ bandgap alteration in graphene oxide with different reduction routes, Appl. Surf.
Sci. 513 (2020) 145396.
PEEK-b composite, indicating that the creation of macropores on the [8] T. Wan, L. Li, M. Guo, Z. Jiao, Z. Wang, Y. Ito, Y. Wan, P. Zhang, Q. Liu,
surfaces of np-HAp/GO/PEEK-p composites does not significantly affect Immobilization via polydopamine of dual growth factors on polyetheretherketone:
their mechanical strength. Preliminary cell studies suggested that the improvement of cell adhesion, proliferation, and osteo-differentiation, J. Mater.
Sci. 54 (2019) 11179–11196.
incorporation of HAp and GO greatly increased cell viability. Further­ [9] Q. Cheng, B. Yuan, X. Chen, X. Yang, H. Lin, X. Zhu, K. Zhang, X. Zhang, Regulation
more, surface macropores on the np-HAp/GO/PEEK-p composites of surface micro/nano structure and composition of polyetheretherketone and their
significantly enhanced cell proliferation as compared to the np-HAp/ influence on the behavior of MC3T3-E1 pre-osteoblasts, J. Mater. Chem. B 7 (2019)
5713–5724.
GO/PEEK-b composite without macropores. This enhancement is
[10] F. Alshehri, M. Alshehri, T. Sumague, A. Niazy, J. Jansen, J. van den Beucken,
attributed to the increased release of bioactive ions from more exten­ H. Alghamdi, Evaluation of peri-implant bone grafting around surface-porous
sively exposed HAp nanoparticles on the surface of np-HAp/GO/PEEK-p dental implants: an in vivo study in a goat model, Materials 12 (2019) 3606–3618.
composites. It was found that the highest cell viability was observed on [11] J.E. Biemond, R. Aquarius, N. Verdonschot, P. Buma, Frictional and bone ingrowth
properties of engineered surface topographies produced by electron beam
np-HAp/GO/PEEK-p2 composite with 396 μm-diameter pores, suggest­ technology, Arch. orthop. traum. su. 131 (2011) 711–718.
ing that the macropore diameter of around 400 μm favored more cell [12] S. Kujala, J. Ryhanen, A. Danilov, J. Tuukkanen, Effect of porosity on the
adhesion and proliferation as compared to the pore diameters of around osteointegration and bone ingrowth of a weight-bearing nickel-titanium bone graft
substitute, Biomaterials 24 (2003) 4691–4697.

7
Z. Huang et al. Materials Science & Engineering C 123 (2021) 111967

[13] K. Vandamme, I. Naert, J.V. Sloten, R. Puers, J. Duyck, Effect of implant surface [36] R. Ma, L. Weng, X. Bao, Z. Ni, S. Song, W. Cai, Characterization of in situ
roughness and loading on peri-implant bone formation, J. Periodontol. 79 (2008) synthesized hydroxyapatite/polyetheretherketone composite materials, Mater.
150–157. Lett. 71 (2012) 117–119.
[14] G. Cai, H. Wang, Y.K. Jung, Z. Xu, J. Zhang, J. He, D. Wang, J.-W. Shin, [37] J.O. Akindoyo, M.D. Beg, S. Ghazali, H.P. Heim, M. Feldmann, Effects of surface
R. Kaewmanee, S. Nabanita, J. Wei, Hierarchically porous surface of PEEK/nMCS modification on dispersion, mechanical, thermal and dynamic mechanical
composite created by femtosecond laser and incorporation of resveratrol exhibiting properties of injection molded PLA-hydroxyapatite composites, Compos. Part A:
antibacterial performances and osteogenic activity in vitro, Compos. Part B: Eng. Appl. S. 103 (2017) 96–105.
186 (2020) 107802. [38] K. Szustakiewicz, B. Stępak, A. Antończak, M. Maj, M. Gazińska, B. Kryszak,
[15] A.S. Brentel, L.M.R. de Vasconcellos, M.V. Oliveira, M.L.d.A. Graca, L.G.O. de J. Pigłowski, Femtosecond laser-induced modification of PLLA/hydroxyapatite
Vasconcellos, C.A.A. Cairo, Y.R. Carvalho, Histomorphometric analysis of pure composite, Polym. Degrad. Stabil. 149 (2018) 152–161.
titanium implants with porous surface versus rough surface, J. Appl. Oral Sci. 14 [39] T. Umemoto, A. Shibata, M. Terakawa, Femtosecond laser irradiation of the
(2006) 213–218. fluorescent molecules-loaded poly (lactic-co-glycolic acid), Appl. Surf. Sci. 417
[16] S. Peng, P. Feng, P. Wu, W. Huang, Y. Yang, W. Guo, C. Gao, C. Shuai, Graphene (2017) 165–169.
oxide as an interface phase between polyetheretherketone and hydroxyapatite for [40] E.A. Sprague, F. Tio, S.H. Ahmed, J.F. Granada, S.R. Bailey, Impact of parallel
tissue engineering scaffolds, Sci. Rep. 7 (2017) 46604. micro-engineered stent grooves on endothelial cell migration, proliferation, and
[17] Z. Huang, Y. Wan, M. Peng, Z. Yang, H. Luo, Incorporating nanoplate-like function: an in vivo correlation study of the healing response in the coronary swine
hydroxyapatite into polylactide for biomimetic nanocomposites via direct melt model, Circ: Cardiovasc. Inte. 5 (2012) 499–507.
intercalation, Compos. Sci. Tech. 185 (2020) 107903. [41] W.R. Walsh, N. Bertollo, C. Christou, D. Schaffner, R.J. Mobbs, Plasma-sprayed
[18] H. Luo, Y. Zhang, D. Gan, Z. Yang, H. Ao, Q. Zhang, F. Yao, Y. Wan, Incorporation titanium coating to polyetheretherketone improves the bone-implant interface,
of hydroxyapatite into nanofibrous PLGA scaffold towards improved breast cancer Spine J. 15 (2015) 1041–1049.
cell behavior, Mater. Chem. Phys. 226 (2019) 177–183. [42] N. Saranya, S. Saravanan, A. Moorthi, B. Ramyakrishna, N. Selvamurugan,
[19] Y. Wan, C. Wu, G. Xiong, G. Zuo, J. Jin, K. Ren, Y. Zhu, Z. Wang, H. Luo, Enhanced osteoblast adhesion on polymeric nano-scaffolds for bone tissue
Mechanical properties and cytotoxicity of nanoplate-like hydroxyapatite/ engineering, J. Biomed. Nanotechnol. 7 (2011) 238–244.
polylactide nanocomposites prepared by intercalation technique, J. Mech. Behav. [43] T. Darmanin, F. Guittard, Recent advances in the potential applications of
Biomed. Mater. 47 (2015) 29–37. bioinspired superhydrophobic materials, J. Mater. Chem. A 2 (2014)
[20] Q. Zhang, J. Dong, M. Peng, Z. Yang, Y. Wan, F. Yao, J. Zhou, C. Ouyang, H. Luo, 16319–16359.
Laser-induced wettability gradient surface on NiTi alloy for potential reduction of [44] R. Ma, D. Guo, Evaluating the bioactivity of a hydroxyapatite-incorporated
thrombosis, Mat. Sci. Eng. C: Mater. 111 (2020) 110847. polyetheretherketone biocomposite, J. Orthop. Surg. Res. 14 (2019) 32–45.
[21] J. Jin, G. Zuo, G. Xiong, H. Luo, Q. Li, C. Ma, D. Li, F. Gu, Y. Ma, Y. Wan, The [45] P. Feng, S. Peng, C. Shuai, C. Gao, W. Yang, S. Bin, A. Min, In situ generation of
inhibition of lamellar hydroxyapatite and lamellar magnetic hydroxyapatite on the hydroxyapatite on biopolymer particles for fabrication of bone scaffolds owning
migration and adhesion of breast cancer cells, J. Mater. Sci. Mater. Med. 25 (2014) bioactivity, ACS Appl. Mater. Interfaces 12 (2020) 46743–46755.
1025–1031. [46] M.J. Yaszemski, R.G. Payne, W.C. Hayes, R. Langer, A.G. Mikos, Evolution of bone
[22] H. Luo, J. Xie, L. Xiong, Z. Yang, G. Zuo, H. Wang, Y. Zhu, Y. Wan, Engineering transplantation: molecular, cellular and tissue strategies to engineer human bone,
photoluminescent and magnetic lamellar hydroxyapatite by facile one-step Se/Gd Biomaterials 17 (1996) 175–185.
dual-dopingJ, Mater. Chem. B 6 (2018) 3515–3521. [47] J.-Y. Rho, L. Kuhn-Spearing, P. Zioupos, Mechanical properties and the hierarchical
[23] G. Zuo, Y. Wan, Y. Zhang, Preparation and characterization of a novel laminated structure of bone, Med. Eng. Phys. 20 (1998) 92–102.
magnetic hydroxyapatite for application on gene delivery, Mater. Lett. 68 (2012) [48] H. Ma, W. Su, Z. Tai, D. Sun, X. Yan, B. Liu, Q. Xue, Preparation and
225–227. cytocompatibility of polylactic acid/hydroxyapatite/graphene oxide
[24] K. Krishnamoorthy, M. Veerapandian, K. Yun, S.-J. Kim, The chemical and nanocomposite fibrous membrane, Chinese Sci. Bull. 57 (2012) 3051–3058.
structural analysis of graphene oxide with different degrees of oxidation, Carbon [49] L. Cao, W. Weng, X. Chen, Y. Ding, Y. Yan, H. Li, H. Zhao, J.-W. Shin, J. Wei, F. Ji,
53 (2013) 38–49. Development of degradable and bioactive composite as bone implants by
[25] M. Naffakh, A.M. Díez-Pascual, Nanocomposite biomaterials based on poly(ether- incorporation of mesoporous bioglass into poly (L-lactide), Compos. Part B: Eng. 77
ether-ketone) (PEEK) and WS2 inorganic nanotubes, J. Mater. Chem. B 2 (2014) (2015) 454–461.
4509–4520. [50] J.S. Fernandes, P. Gentile, R.A. Pires, R.L. Reis, P.V. Hatton, Multifunctional
[26] M. Kuo, C. Tsai, J. Huang, M. Chen, PEEK composites reinforced by nano-sized SiO2 bioactive glass and glass-ceramic biomaterials with antibacterial properties for
and Al2O3 particulates, Mater. Chem. Phys. 90 (2005) 185–195. repair and regeneration of bone tissue, Acta Biomater. 59 (2017) 2–11.
[27] S. Chaiyakun, N. Witit-Anun, N. Nuntawong, P. Chindaudom, S. Oaew, [51] C. Shuai, W. Yang, P. Feng, S. Peng, H. Pan, Accelerated degradation of HAP/PLLA
C. Kedkeaw, P. Limsuwan, Preparation and characterization of graphene oxide bone scaffold by PGA blending facilitates bioactivity and osteoconductivity,
nanosheets, Procedia Eng. 32 (2012) 759–764. Bioactive Mater 6 (2021) 490–502.
[28] H. Fujimori, H. Toya, K. Ioku, S. Goto, M. Yoshimura, In situ observation of defects [52] N.A. Aval, R. Emadi, A. Valiani, M. Kharaziha, M. Karimipour, R. Rahbarghazi,
in hydroxyapatite up to 1200 ◦ C by ultraviolet raman spectroscopy, Chem. Phys. Nano-featured poly (lactide-co-glycolide)-graphene microribbons as a promising
Lett. 325 (2000) 383–388. substrate for nerve tissue engineering, Compos. Part B: Eng. 173 (2019) 106863.
[29] H. Luo, W. Li, D. Ji, G. Zuo, G. Xiong, Y. Zhu, L. Li, M. Han, C. Wu, Y. Wan, One- [53] V. Sgarminato, C. Tonda-Turo, G. Ciardelli, Reviewing recently developed
step exfoliation and surface modification of lamellar hydroxyapatite by technologies to direct cell activity through the control of pore size: from the macro-
intercalation of glucosamine, Mater. Chem. Phys. 2016 (2016) 1–6. to the nanoscale, J. Biomed. Mater. Res. B 108 (2020) 1176–1185.
[30] G. Zuo, Y. Wan, X. Meng, Q. Zhao, K. Ren, S. Jia, J. Wang, Synthesis and [54] S. Yamane, N. Iwasaki, Y. Kasahara, K. Harada, T. Majima, K. Monde, S. Nishimura,
characterization of a lamellar hydroxyapatite/DNA nanohybrid, Mater. Chem. A. Minami, Effect of pore size on in vitro cartilage formation using chitosan-based
Phys. 126 (2011) 470–475. hyaluronic acid hybrid polymer fibers, J. Biomed. Mater. Res. A 81 (2007)
[31] M. Raucci, D. Giugliano, A. Longo, S. Zeppetelli, G. Carotenuto, L. Ambrosio, 586–593.
Comparative facile methods for preparing graphene oxide-hydroxyapatite for bone [55] J. Klawitter, J. Bagwell, A. Weinstein, B. Sauer, J. Pruitt, An evaluation of bone
tissue engineering, J. Tissue Eng. Regen. M. 11 (2017) 2204–2216. growth into porous high density polyethylene, J. Biomed. Mater. Res. 10 (1976)
[32] X.-Z. Xue, J.-Y. Zhang, D. Zhou, J.-K. Liu, In-situ bonding technology and excellent 311–323.
anticorrosion activity of graphene oxide/hydroxyapatite nanocomposite pigment, [56] Y. Zhao, K. Tan, Y. Zhou, Z. Ye, W.-S. Tan, A combinatorial variation in surface
Dyes Pigments 160 (2019) 109–118. chemistry and pore size of three-dimensional porous poly (epsilon-caprolactone)
[33] M. Ramadas, G. Bharath, N. Ponpandian, A. Ballamurugan, Investigation on scaffolds modulates the behaviors of mesenchymal stem cells, Mat, Sci. Eng. C:
biophysical properties of hydroxyapatite/graphene oxide (HAp/GO) based binary Mater. 59 (2016) 193–202.
nanocomposite for biomedical applications, Mater. Chem. Phys. 199 (2017) [57] T.H. Kim, S.H. Oh, E.B. Kwon, J.Y. Lee, J.H. Lee, In vitro evaluation of osteogenesis
179–184. and myogenesis from adipose-derived stem cells in a pore size gradient scaffold,
[34] R. Ma, L. Weng, L. Fang, Z. Luo, S. Song, Structure and mechanical performance of Macromol. Res. 21 (2013) 878–885.
in situ synthesized hydroxyapatite/polyetheretherketone nanocomposite materials, [58] S.H. Oh, T.H. Kim, G.I. Im, J.H. Lee, Investigation of pore size effect on
J. Sol-Gel Sci. Techn. 62 (2012) 52–56. chondrogenic differentiation of adipose stem cells using a pore size gradient
[35] Y.-S. Chen, J.-H.C. Lin, Y.-R. Wu, C.-W. Chang, K.-C. Chang, C.-C. Chen, C.-H. Chen, scaffold, Biomacromolecules 11 (2010) 1948–1955.
W.-C. Chen, Characterizing the differentiation of osteoprogenitor cells on surface
modified polyether-ether-ketone, Surf. Coat. Tech. 350 (2018) 904–912.