a ph-sensitive self-healing coating for biodegradable magnesium … · 2019-11-07 · self-healing...

Acta Biomaterialia 98 (2019) 160–173

Full length article

A pH-sensitive self-healing coating for biodegradable magnesiumimplantsq

Pan Xiong a, JiangLong Yan a, Pei Wang a, ZhaoJun Jia b, Wenhao Zhou a, Wei Yuan c, Yangyang Li a, Yang Liu c,Yan Cheng a,⇑, Dafu Chen d,⇑, Yufeng Zheng a,c,⇑aBiomed-X Center, Academy for Advanced Interdisciplinary Studies, Peking University, Beijing 100871, ChinabDepartment of Orthopaedic and Traumatology, The University of Hong Kong, 21 Sassoon Road, Pokfulam, Hong Kong, ChinacDepartment of Materials Science and Engineering, College of Engineering, Peking University, Beijing 100871, Chinad Laboratory of Bone Tissue Engineering, Beijing Research Institute of Traumatology and Orthopaedics, Beijing JiShuiTan Hospital, Beijing 100035, China

a r t i c l e i n f o

Article history:Received 30 November 2018Received in revised form 14 March 2019Accepted 22 April 2019Available online xxxx

Keywords:Biomedical Mg alloySelf-healingSilk fibroinBiodegradationOsteogenic activity

a b s t r a c t

Self-healing coatings have attracted attention on surface modification of magnesium alloys, as it canrecover the barrier ability of the coatings from corrosion attack. Nevertheless, previous works on thisaspect are not suitable for biomedical magnesium alloys owing to the lack of biocompatibility. In thisstudy, we fabricated a self-healing coating on biomedical Mg-1Ca alloy by compositing silk fibroin andK3PO4. PO4

3� ions act as corrosion inhibitor, while K3+ ions help to regulate the secondary structures ofsilk fibroin. The scratch test, scanning vibrating electrode technique (SVET), and electrochemical impe-dance spectroscopy (EIS) provide comprehensive results, confirming the pH-sensitive self-healing capac-ity of the composite coating. Moreover, cells’ (MC3T3-E1) multiple responses including spreading,adhesion, proliferation, and differentiation illustrate the preferable biocompatibility as well as the osteo-genic activity of the coating. These primary findings might open new opportunities in the exploration ofself-healing coatings on biomedical magnesium alloys.

Statement of significance

Biomedical magnesium alloys surface modifications have been studied for years, which however thebiomedical self-healing coatings were rarely involved. In this work, silk fibroin and phosphate (K3PO4)were composited to fabricate coating on biomedical magnesium alloys. The coating not only ownedthe self-healing ability with pH sensitivity, but also endowed the substrate preferable corrosion resis-tance as well as osteogenic activity. This work gives a new insight into surface modification for biomed-ical Mg alloys.

� 2019 Acta Materialia Inc. Published by Elsevier Ltd. All rights reserved.

1. Introduction

Recently, the temporary implant concept of magnesium (Mg)and its alloys has attracted increasing attention on orthopedics ofspecial clinical applications, but the speedy corrosion rate still lim-its their practical implementation [1,2]. Rapid degradation in phys-iological ambient can lead to absence of mechanical integrity, local

alkalosis, and release of copious hydrogen [3,4]. Surface modifica-tion is a well-known technology to efficiently improve the corro-sion resistance of Mg alloys [5,6]. Presently, various surfacemodification methods have been developed, such as phosphatecoating [7,8], microarc oxidation (MAO) [9,10], fluoride conversionfilm [11,12], and polymer coatings [13–15]. These methods mainlyimprove the corrosion resistance of Mg alloys by playing the role ofbarriers to corrosion electrolyte.

Self-healing coatings have been studied in the industry inrecent years [16]. For these coatings, once the defects appear onthe surface and the protective barrier is destroyed, they canrecover the protection without external help. The self-healing coat-ings are constructed by two parts, namely, skeleton coatingsderived from polymers such as PSS [17], polyester [18], and PEI

https://doi.org/10.1016/j.actbio.2019.04.0451742-7061/� 2019 Acta Materialia Inc. Published by Elsevier Ltd. All rights reserved.

q Part of the Special Issue associated with the 10th International Conference onBiodegradable Metals, 10th Biometal 2018, held at the University of Oxford, 26–31Aug. 2018, organized by Professors Diego Mantovani and Frank Witte.⇑ Corresponding authors at: Department of Materials Science and Engineering,

College of Engineering, Peking University, Beijing 100871, China (Y. Zheng).E-mail addresses: [email protected] (Y. Cheng), [email protected]

(D. Chen), [email protected] (Y. Zheng).

Acta Biomaterialia xxx (xxxx) xxx

Contents lists available at ScienceDirect

Acta Biomaterialia

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

Please cite this article as: P. Xiong, J. Yan, P. Wang et al., A pH-sensitive self-healing coating for biodegradable magnesium implants, Acta Biomaterialia,https://doi.org/10.1016/j.actbio.2019.04.045

25 April 2019

http://crossmark.crossref.org/dialog/?doi=10.1016/j.actbio.2019.04.045&domain=pdf

P.Xiongetal./ActaBiomaterialia98(2019)160–173 161

Full length article

A pH-sensitive self-healing coating for biodegradable magnesiumimplantsq

Pan Xiong a, JiangLong Yan a, Pei Wang a, ZhaoJun Jia b, Wenhao Zhou a, Wei Yuan c, Yangyang Li a, Yang Liu c,Yan Cheng a,⇑, Dafu Chen d,⇑, Yufeng Zheng a,c,⇑aBiomed-X Center, Academy for Advanced Interdisciplinary Studies, Peking University, Beijing 100871, ChinabDepartment of Orthopaedic and Traumatology, The University of Hong Kong, 21 Sassoon Road, Pokfulam, Hong Kong, ChinacDepartment of Materials Science and Engineering, College of Engineering, Peking University, Beijing 100871, Chinad Laboratory of Bone Tissue Engineering, Beijing Research Institute of Traumatology and Orthopaedics, Beijing JiShuiTan Hospital, Beijing 100035, China

a r t i c l e i n f o

Article history:Received 30 November 2018Received in revised form 14 March 2019Accepted 22 April 2019Available online xxxx

Keywords:Biomedical Mg alloySelf-healingSilk fibroinBiodegradationOsteogenic activity

a b s t r a c t

Self-healing coatings have attracted attention on surface modification of magnesium alloys, as it canrecover the barrier ability of the coatings from corrosion attack. Nevertheless, previous works on thisaspect are not suitable for biomedical magnesium alloys owing to the lack of biocompatibility. In thisstudy, we fabricated a self-healing coating on biomedical Mg-1Ca alloy by compositing silk fibroin andK3PO4. PO4

3� ions act as corrosion inhibitor, while K3+ ions help to regulate the secondary structures ofsilk fibroin. The scratch test, scanning vibrating electrode technique (SVET), and electrochemical impe-dance spectroscopy (EIS) provide comprehensive results, confirming the pH-sensitive self-healing capac-ity of the composite coating. Moreover, cells’ (MC3T3-E1) multiple responses including spreading,adhesion, proliferation, and differentiation illustrate the preferable biocompatibility as well as the osteo-genic activity of the coating. These primary findings might open new opportunities in the exploration ofself-healing coatings on biomedical magnesium alloys.

Statement of significance

Biomedical magnesium alloys surface modifications have been studied for years, which however thebiomedical self-healing coatings were rarely involved. In this work, silk fibroin and phosphate (K3PO4)were composited to fabricate coating on biomedical magnesium alloys. The coating not only ownedthe self-healing ability with pH sensitivity, but also endowed the substrate preferable corrosion resis-tance as well as osteogenic activity. This work gives a new insight into surface modification for biomed-ical Mg alloys.

� 2019 Acta Materialia Inc. Published by Elsevier Ltd. All rights reserved.

1. Introduction

Recently, the temporary implant concept of magnesium (Mg)and its alloys has attracted increasing attention on orthopedics ofspecial clinical applications, but the speedy corrosion rate still lim-its their practical implementation [1,2]. Rapid degradation in phys-iological ambient can lead to absence of mechanical integrity, local

alkalosis, and release of copious hydrogen [3,4]. Surface modifica-tion is a well-known technology to efficiently improve the corro-sion resistance of Mg alloys [5,6]. Presently, various surfacemodification methods have been developed, such as phosphatecoating [7,8], microarc oxidation (MAO) [9,10], fluoride conversionfilm [11,12], and polymer coatings [13–15]. These methods mainlyimprove the corrosion resistance of Mg alloys by playing the role ofbarriers to corrosion electrolyte.

Self-healing coatings have been studied in the industry inrecent years [16]. For these coatings, once the defects appear onthe surface and the protective barrier is destroyed, they canrecover the protection without external help. The self-healing coat-ings are constructed by two parts, namely, skeleton coatingsderived from polymers such as PSS [17], polyester [18], and PEI

https://doi.org/10.1016/j.actbio.2019.04.0451742-7061/� 2019 Acta Materialia Inc. Published by Elsevier Ltd. All rights reserved.

q Part of the Special Issue associated with the 10th International Conference onBiodegradable Metals, 10th Biometal 2018, held at the University of Oxford, 26–31Aug. 2018, organized by Professors Diego Mantovani and Frank Witte.⇑ Corresponding authors at: Department of Materials Science and Engineering,

College of Engineering, Peking University, Beijing 100871, China (Y. Zheng).E-mail addresses: [email protected] (Y. Cheng), [email protected]

(D. Chen), [email protected] (Y. Zheng).

Acta Biomaterialia xxx (xxxx) xxx

Contents lists available at ScienceDirect

Acta Biomaterialia

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


[19], and corrosion inhibitors such as Ce2+ ion [20], 8-hydroxyquinoline [21,22], and benzotriazole [23,24]. These meth-ods are efficient for industrial magnesium alloys but not suitablefor biomedical materials, as both skeleton coatings and corrosioninhibitors lack biosafety and biocompatibility.

Silk fibroin is a typical natural protein extracted from the silk-worm moth Bombyx mori [25]. The main elements of silk fibroinare repeats of the GAGAGS motif, composed of a heavy chain(325 kDa) and a light chain (25 kDa), which are connected by adisulfide bond and complexed by a small glycoprotein, P25(30 kDa) [26]. The most significant feature of silk fibroin is the con-trollable secondary structures. Generally, the secondary structuresof silk fibroin contain random coils, a-helices, b-sheets, and b-turns, which can be adjusted by metal ions [27], solvent [28],post-treatment [29], shearing force [30], etc. The proportion of var-ious secondary structures is related to mechanical properties, sta-bility, wettability, degradability, and biocompatibility, which arekey factors for silk fibroin in biomaterial applications [31]. Further-more, silk fibroin has been investigated for the loading capacity,especially for drugs [32,33], hydroxyapatite [34,35], and BMP-2[36,37]. Silk fibroin has limited inflammatory responses whencompared with other polymers evaluated in biomaterials such asPLA, PGA, and PLGA, which degrade into acid monomers [38].

In the present work, inspired by the self-healing coatings onindustrial magnesium alloys, we establish self-healing coating onbiomedical Mg-1Ca alloy by compositing silk fibroin and K3PO4.Mg-1Ca alloy is a newly designed promising magnesium alloy forbone implants because of the preferable mechanical propertiesand release of Mg2+ and Ca2+ ions during degradation that can pro-mote bone healing [39]. However, the corrosion rate of the Mg-1Caalloy is too rapid for application in the clinic. Silk fibroin acts as theskeleton coating, while the PO4

3� ions serve as the corrosion inhibi-tor because of the lower solubility of the formed Mg3(PO4)2. Fur-thermore, silk fibroin is known as an excellent biomaterial withosteogenic activity, and the PO4

3� ion is the inorganic ingredientof bone. Thus, the composite coating is expected to possess self-healing ability as well as osteogenic activity, which can be a newoption for surface modification of magnesium alloys.

2. Materials and methods

2.1. Substrate preparation

Mg-1Ca alloy was extruded according to our previous work[40]. The grain size of the Mg-1Ca alloy was approximately70.5 ± 1.5 nm, and its roughness was 51.3 ± 2.5 nm. It was cut intodiscs (U = 12 mm, h = 2 mm); ground using 2000-grit SiC papers;then ultrasonically washed in acetone, ethanol, and deionizedwater, respectively; and dried in cold air sequentially.

2.2. Fluoride pretreatment

The fluoride treatment (step 1 in Fig. 1) was processed by twosteps. Briefly, the ground samples were immersed in boiling NaOH(5 M) solution for 3 h and then rinsed with deionized water andimmersed in HF (40%) solution for 6 h at 60 �C. The as-preparedsamples were rinsed and dried at room temperature.

2.3. Self-healing coating formulation

The silk fibroin solution was prepared following the previousprotocol [25]. Cocoons were boiled in 0.02 M Na3CO3 aqueous solu-tion for 30 min and then rinsed with DI water to extract sericin.After drying in room temperature overnight, the degummedfibroin was dissolved in 9.3 M LiBr solution for 4 h at 60 �C.

Subsequently, the solution was dialyzed against DI water for3 days. Finally, silk fibroin with a preliminary concentration ofapproximately 8% (m/v) was obtained.

The obtained silk fibroin solution was diluted, and the K3PO4

solution was added to the silk fibroin. The final concentration ofsilk fibroin was 2 wt%, and the concentration of K3PO4 was 0.2 M.A spin-coater (Model KW-4A Spin Coater, Siyouyen�, Beijing,China), with a low rotation speed of 500 rpm for 12 s and a highrotation speed of 4000 rpm for 10 s at room temperature, wasemployed to fabricate the coating by the spin-coating method.Fifty microliters of the solution was added to the specimen for acycle, and this was repeated five times for each side of the sample.The target coatings were formed by three silk fibroin and K3PO4

composite coatings and two pure silk fibroin coatings, and thecoated samples were denoted as ‘‘Silk-KP” (shown in Fig. 1 Steps2 and 3). The pure silk fibroin coated samples were used as thecontrol and denoted as ‘‘Pure Silk” (Fig. 1 Step 2’). The fabricationprocess is illustrated in Fig. 1.

2.4. Coating characterization

To evaluate the structure of K3PO4-treated silk fibroin, the sur-face coating obtained in Step 2 (denoted as ‘‘Silk-K3PO4 coating”)and Step 2’ (‘‘Pure Silk” as control) of Fig. 1 was characterized.The chemical composition and states of the designed coatings weredetermined by X-ray photoelectron spectroscopy (XPS; AXIS Ultra,Kratos) with Al Ka irradiation (ht = 1486.71 eV). The surface func-tional groups of the coatings were recognized by Attenuated TotalReflectance-Fourier Transform Infrared analysis (ATR-FTIR, ThermoFisher Scientific), with spectra recording from 4000 cm�1 to400 cm�1. Peak Fit software was used to fit and calculate the sec-ondary structures of silk fibroin.

The Nanoindenter System (Hysitron, USA) was used to measurethe adhesion resistance and mechanical properties (hardness,Young’s modulus, and wear resistance). Adhesion resistance wasmeasured by the ramp-load scratch procedure with a diamondcone tip (r = 1 mm) under a loading force of 0 mN to 2000 mN. Wearresistance was measured by the vertical-load scratch procedurewith the diamond cone tip (r = 1 mm) under a loading force of100 mN. Hardness and Young’s modulus were measured by thevertical-load scratch procedure with a diamond triangular pyramidtip (r = 50 nm) poking at a depth of 60 nm. Five duplicates weremeasured in each test.

Fig. 1. Schematic of the Silk-PK fabrication process (the scheme is not in real scale).

2 P. Xiong et al. / Acta Biomaterialia xxx (xxxx) xxx


162 P.Xiongetal./ActaBiomaterialia98(2019)160–173

2.5. In vitro corrosion evaluation

Hank’s solution (NaCl 8.00 g/L, KCl 0.40 g/L, CaCl2 0.14 g/L,NaHCO3 0.35 g/L, MgSO4�7H2O 0.20 g/L, Na2HPO4�12H2O, 0.12 g/L,KH2PO4 0.06 g/L, pH = 7.4) was adopted for immersion experimentaccording to ASTM-G31-72 at 37 �C. The hydrogen evolution vol-ume and pH value of the solution were recorded for 14 days ofimmersion. After immersion for 7 days, samples in Hank’s solutionwere taken out, rinsed with deionized water, and air-dried,sequentially. Changes in surface morphologies were assessed bySEM (FE-SEM, S4800, Hitachi), and the main elements of the sur-face were detected by EDX, XPS, and ATR-FTIR. Three duplicateswere taken for each group.

2.6. Self-healing capability

2.6.1. Scratch testsArtificial scratches were measured using the Nanoindenter Sys-

tem (Hysitron, USA) with a diamond cone tip (radius of 1 lm). Toobtain a similar scratch on each sample, the tip was plunged intothe surface and loaded with a constant force of 100 lN. The 3-dimensional size of the scratches was recorded and imaged. Thesamples with the scratches were immersed in Hank’s solution at37 �C. After immersion for 24 h, the size of the scratches wasrecorded and imaged again. Furthermore, the elements in scratchesafter immersion were detected by EDX.

2.6.2. Scanning vibrating electrode technique (SVET) measurementsSVET measurements were made on M370 Scanning Electro-

chemical Workstation (Ametek, USA). Silk and Silk-KP sampleswith artificial defects were fixed on an epoxy resin holder andimmersed in Hank’s solution. The measurement was controlledusing ASET software, and the system recorded the real-time vectorcurrent density of the assigned surface. The Pt-blackened electrodetip (15 lm diameter) was situated on the tested surface at a dis-tance of 100 ± 2 lm. The electric current densities of the exposedarea (1.5 � 1.5 mm2) were scanned with a step size of 60 lm ona lattice of 25 � 25 points and presented in the form of 3D maps.Scans were performed after 40 min of immersion and automati-cally collected every 40 min during the experiments. The sampleswere monitored for 3 h.

2.6.3. Electrochemical impedance spectroscopy (EIS) measurementsThe EIS measurement was performed using an electrochemistry

workstation (PGSTAT 302N, Metrohm Autolab) connected to astandard three-electrode cell with a saturated calomel electrode(SCE) as the reference electrode and a platinum piece as the coun-ter electrode in Hank’s solution at 37 �C. The samples immersed inHank’s solution at various time points served as the working elec-trode, and the area exposed to the electrolyte was 0.20 cm2. Thestudy was conducted at OCP in the frequency range of 100 mHzto 100 kHz. The EIS plots that were generated were analyzed usingNova 2.1 software, and they were best-fitted to the appropriateequivalent circuit (EC) models. An average of three measurementswas taken for each group.

2.6.4. pH stimuli-responsive abilityTo evaluate the effect of the pH value on the designed coating,

the Silk-KP was coated onto glasses (the same size as that of theMg-1Ca alloy sample) by the same method to exclude the impactof the substrate. DI water of pH values 7.4 and 10.0, which wereadjusted by NaOH (0.1 mol/L), was employed as the immersionmedium to exclude the influence of other substances. Sampleswere immersed in the as-prepared DI water with a solutionvolume-to-surface area of the sample ratio of 20 ml/cm2. Atpredetermined time points, the entire volume of the solution was

collected and refilled with fresh DI water of pH values 7.4 and10.0. The concentration of the element P (derived from PO4

3�) inthe solution was detected by inductively coupled plasma massspectrometry (ICP-MS; Agilent 7700X, USA). An average of at leastfive parallel samples was taken for each group.

2.7. Cytocompatibility and osteogenic activity

2.7.1. Cell culture and seedingFor in vitro biocompatibility measurements, mouse osteoblast-

like cells (MC3T3-E1) were adopted and cultured in a-MEM (min-imum essential medium alpha, HyClone, Beijing) containing 10%FBS (fetal bovine serum, HyClone, Beijing) and 1% penicillin andstreptomycin (HyClone, Beijing) in a humidified incubator with5% CO2 at 37 �C. The contents of a-MEM are listed in Table S1.The extracts of each sample were obtained following the instruc-tion of ISO10993-12. For all cell response experiments, a-MEMalone containing 10% FBS was added per well into TCPS as the con-trol. The culture medium was refreshed every two days during theculture period. Each group had at least five duplicate wells in eachevaluation.

2.7.2. Cell spreading morphologyCells were cultured in extracts or on the samples directly for 6 h

and 24 h, respectively. At both these time points, on the one hand,the cells were sequentially rinsed with PBS, fixed using 4% (w/v)paraformaldehyde for 10 min, permeabilized with 0.1% (v/v) TritonX-100 (Sigma) for 10 min, stained with 1.0% (v/v) FITC-phalloidin(Sigma) for 30 min and 1 mg/mL DAPI (Sigma) for 5 min, andimaged using an LSCM (using a laser scanning confocal microscope,Nikon ALR-SI; identical apparatus throughout the experiments); onthe other hand, the samples were rinsed with PBS, fixed with 2.5%glutaraldehyde solution for 2 h at room temperature, dehydratedin a gradient ethanol/DI water (ethanol concentrations of 50%,60%, 70%, 80%, 90%, 95%, and 100%), and observed using an SEM(FE-SEM, S4800, Hitachi).

2.7.3. Cell viabilityA mitochondrial activity-based cell counting kit (CCK-8,

Dojindo, Japan) was adopted to determine cell viability. Cells werecultured for 1, 3, and 5 days in extracts for measurements, and thea-MEM with 10% FBS was used as the negative control, while a-MEM containing 10% DMSO (dimethyl sulfoxide) was used as thepositive control. Cells were cultured in the 96-well plate, and 5duplicates were taken for each sample. At scheduled time points,10% CCK-8 solution was added to the medium and incubated foranother 2 h to generate formazan. The OD values ([A]) at 450 nmwere read on a microplate spectrophotometer (Bio-Rad, USA). Cellviability was calculated using the formula

%cell viability ¼ A½ �Sample� A½ �PositiveA½ �Negative� A½ �Positive

� �� 100%

2.7.4. Osteogenic differentiation studiesALP activity was measured by the colorimetrical production of

p-nitrophenol (p-NP) through p-nitrophenyl phosphate (p-NPP)/endogenous ALP enzymatic reaction (Jiancheng, Nanjing, China).After culturing for 7 days and 14 days, the cells were lysed in 1%Triton X-100 for 1 h. Then, 30 mL of the cell lysates was culturedwith carbonate buffer solution (50 mL) and substrate solution (4-amino-antipyrine, 50 mL) for 15 min at 37 �C. Afterwards, a chro-mogenic agent (potassium ferricyanide, 150 mL) was added intothe mixed solution. The absorbance of the mixed solution was readat 520 nm using the microplate spectrophotometer (Bio-Rad, USA).Moreover, the ALP activity was normalized using the Micro BCA

P. Xiong et al. / Acta Biomaterialia xxx (xxxx) xxx 3



2.5. In vitro corrosion evaluation

Hank’s solution (NaCl 8.00 g/L, KCl 0.40 g/L, CaCl2 0.14 g/L,NaHCO3 0.35 g/L, MgSO4�7H2O 0.20 g/L, Na2HPO4�12H2O, 0.12 g/L,KH2PO4 0.06 g/L, pH = 7.4) was adopted for immersion experimentaccording to ASTM-G31-72 at 37 �C. The hydrogen evolution vol-ume and pH value of the solution were recorded for 14 days ofimmersion. After immersion for 7 days, samples in Hank’s solutionwere taken out, rinsed with deionized water, and air-dried,sequentially. Changes in surface morphologies were assessed bySEM (FE-SEM, S4800, Hitachi), and the main elements of the sur-face were detected by EDX, XPS, and ATR-FTIR. Three duplicateswere taken for each group.

2.6. Self-healing capability

2.6.1. Scratch testsArtificial scratches were measured using the Nanoindenter Sys-

tem (Hysitron, USA) with a diamond cone tip (radius of 1 lm). Toobtain a similar scratch on each sample, the tip was plunged intothe surface and loaded with a constant force of 100 lN. The 3-dimensional size of the scratches was recorded and imaged. Thesamples with the scratches were immersed in Hank’s solution at37 �C. After immersion for 24 h, the size of the scratches wasrecorded and imaged again. Furthermore, the elements in scratchesafter immersion were detected by EDX.

2.6.2. Scanning vibrating electrode technique (SVET) measurementsSVET measurements were made on M370 Scanning Electro-

chemical Workstation (Ametek, USA). Silk and Silk-KP sampleswith artificial defects were fixed on an epoxy resin holder andimmersed in Hank’s solution. The measurement was controlledusing ASET software, and the system recorded the real-time vectorcurrent density of the assigned surface. The Pt-blackened electrodetip (15 lm diameter) was situated on the tested surface at a dis-tance of 100 ± 2 lm. The electric current densities of the exposedarea (1.5 � 1.5 mm2) were scanned with a step size of 60 lm ona lattice of 25 � 25 points and presented in the form of 3D maps.Scans were performed after 40 min of immersion and automati-cally collected every 40 min during the experiments. The sampleswere monitored for 3 h.

2.6.3. Electrochemical impedance spectroscopy (EIS) measurementsThe EIS measurement was performed using an electrochemistry

workstation (PGSTAT 302N, Metrohm Autolab) connected to astandard three-electrode cell with a saturated calomel electrode(SCE) as the reference electrode and a platinum piece as the coun-ter electrode in Hank’s solution at 37 �C. The samples immersed inHank’s solution at various time points served as the working elec-trode, and the area exposed to the electrolyte was 0.20 cm2. Thestudy was conducted at OCP in the frequency range of 100 mHzto 100 kHz. The EIS plots that were generated were analyzed usingNova 2.1 software, and they were best-fitted to the appropriateequivalent circuit (EC) models. An average of three measurementswas taken for each group.

2.6.4. pH stimuli-responsive abilityTo evaluate the effect of the pH value on the designed coating,

the Silk-KP was coated onto glasses (the same size as that of theMg-1Ca alloy sample) by the same method to exclude the impactof the substrate. DI water of pH values 7.4 and 10.0, which wereadjusted by NaOH (0.1 mol/L), was employed as the immersionmedium to exclude the influence of other substances. Sampleswere immersed in the as-prepared DI water with a solutionvolume-to-surface area of the sample ratio of 20 ml/cm2. Atpredetermined time points, the entire volume of the solution was

collected and refilled with fresh DI water of pH values 7.4 and10.0. The concentration of the element P (derived from PO4

3�) inthe solution was detected by inductively coupled plasma massspectrometry (ICP-MS; Agilent 7700X, USA). An average of at leastfive parallel samples was taken for each group.

2.7. Cytocompatibility and osteogenic activity

2.7.1. Cell culture and seedingFor in vitro biocompatibility measurements, mouse osteoblast-

like cells (MC3T3-E1) were adopted and cultured in a-MEM (min-imum essential medium alpha, HyClone, Beijing) containing 10%FBS (fetal bovine serum, HyClone, Beijing) and 1% penicillin andstreptomycin (HyClone, Beijing) in a humidified incubator with5% CO2 at 37 �C. The contents of a-MEM are listed in Table S1.The extracts of each sample were obtained following the instruc-tion of ISO10993-12. For all cell response experiments, a-MEMalone containing 10% FBS was added per well into TCPS as the con-trol. The culture medium was refreshed every two days during theculture period. Each group had at least five duplicate wells in eachevaluation.

2.7.2. Cell spreading morphologyCells were cultured in extracts or on the samples directly for 6 h

and 24 h, respectively. At both these time points, on the one hand,the cells were sequentially rinsed with PBS, fixed using 4% (w/v)paraformaldehyde for 10 min, permeabilized with 0.1% (v/v) TritonX-100 (Sigma) for 10 min, stained with 1.0% (v/v) FITC-phalloidin(Sigma) for 30 min and 1 mg/mL DAPI (Sigma) for 5 min, andimaged using an LSCM (using a laser scanning confocal microscope,Nikon ALR-SI; identical apparatus throughout the experiments); onthe other hand, the samples were rinsed with PBS, fixed with 2.5%glutaraldehyde solution for 2 h at room temperature, dehydratedin a gradient ethanol/DI water (ethanol concentrations of 50%,60%, 70%, 80%, 90%, 95%, and 100%), and observed using an SEM(FE-SEM, S4800, Hitachi).

2.7.3. Cell viabilityA mitochondrial activity-based cell counting kit (CCK-8,

Dojindo, Japan) was adopted to determine cell viability. Cells werecultured for 1, 3, and 5 days in extracts for measurements, and thea-MEM with 10% FBS was used as the negative control, while a-MEM containing 10% DMSO (dimethyl sulfoxide) was used as thepositive control. Cells were cultured in the 96-well plate, and 5duplicates were taken for each sample. At scheduled time points,10% CCK-8 solution was added to the medium and incubated foranother 2 h to generate formazan. The OD values ([A]) at 450 nmwere read on a microplate spectrophotometer (Bio-Rad, USA). Cellviability was calculated using the formula

%cell viability ¼ A½ �Sample� A½ �PositiveA½ �Negative� A½ �Positive

� �� 100%

2.7.4. Osteogenic differentiation studiesALP activity was measured by the colorimetrical production of

p-nitrophenol (p-NP) through p-nitrophenyl phosphate (p-NPP)/endogenous ALP enzymatic reaction (Jiancheng, Nanjing, China).After culturing for 7 days and 14 days, the cells were lysed in 1%Triton X-100 for 1 h. Then, 30 mL of the cell lysates was culturedwith carbonate buffer solution (50 mL) and substrate solution (4-amino-antipyrine, 50 mL) for 15 min at 37 �C. Afterwards, a chro-mogenic agent (potassium ferricyanide, 150 mL) was added intothe mixed solution. The absorbance of the mixed solution was readat 520 nm using the microplate spectrophotometer (Bio-Rad, USA).Moreover, the ALP activity was normalized using the Micro BCA



Protein Kit (Themo Fisher Scientific). The results were expressed asmM production of p-NP by each gram of protein (mM p-NP/gprot).Additionally, the cells were fixed in 4% PFA and stained with BCIP/NBT ALP Color Development Kit (Beyotime, China) for qualitativeimaging. Images were captured using an inverted microscope(Olympus IX73, Japan).

For collagen secretion and ECM mineralization assays, cells thatwere cultured for 28 days were rinsed with PBS and fixed in 4% PFAfor 15 min. For staining, the cells were stained in Sirius Red (SR,0.1%, Sigma) overnight or Alizarin Red S (ARS, Sigma, 2%,pH = 4.3) for 10 min at 4 �C and washed until no more color

appeared. Optical images for each well were acquired using anoptical microscope with a camera. Moreover, the stain was elutedin 200 mL of 50% 0.2 M NaOH/methanol solution or 10% cetylpyri-dinium chloride, and the absorbance of the resulting solutionwas measured at 570 nm for quantity.

2.8. Statistical analysis

Quantitative data, presented as means ± standard deviations,were obtained for n � 3. Statistical analysis was performed using

Fig. 2. (A1-A3) XPS results comprising the survey spectra and co-responding core-level spectra for C 1s of the Silk-K3PO4 coating; (B1) ATR-FTIR spectra of the Silk and Silk-K3PO4 coating; (B2) and (B3) Absorbance spectra, amide I, deduced after Fourier self-deconvolution of the Silk and Silk-K3PO4 coating, respectively; (C) Illustrative diagrams ofthe secondary structural changes of silk fibroin composited with K+ ions. The scheme is not in real scale.




one-way analysis of variance (ANOVA), and the significance wasregarded at p value < 0.05.

3. Results

3.1. Chemical composition of the surface coating

The chemical and structural changes after K3PO4 was added tothe silk fibroin were determined by XPS and ATR-FTIR. As shownin Fig. 2 (A1-A2), the survey spectrum of XPS results showed thatthe main elements on the surfaces of both the Silk-K3PO4 coatingand the Silk were C, N, and O, which were the main elements of silkfibroin. Additionally, the P element was detected on the Silk-K3PO4

coating, which confirmed the addition of K3PO4. Furthermore, theC1s spectrum for the Silk-K3PO4 coating (Fig. 2(A3)) showed char-acteristic peaks at 285 (C–C), 286 (C–H), and 288 (C@O), corre-sponding to different chemical environments of silk fibroin [41,42].

As shown in Fig. 2(B1), the functional groups of both the Silk-K3PO4 coating and the Silk were detected by FTIR, and two peaksat 1655 cm�1 and 1540 cm�1 were found for these two samples,respectively. It was attributed to the amide I (N–H deformation)and amide II (C–N stretching) [43], respectively, which were char-acteristic peaks of silk fibroin. Moreover, the peaks at 1065 cm�1,982 cm�1, and 824 cm�1 were assigned to v3 vibration peak ofPO4

3� [44], which was obtained for the Silk-K3PO4 coating only.To further realize the secondary structural changes of silk fibroinafter K3PO4 was added to the silk fibroin, the amide I region(1590–1700 cm�1) was further characterized by deconvolution ofthe ATR-FTIR spectra. The proportion of the main secondary struc-tures was calculated, and the data are given in Table 1. The contentof b-sheets was more than 30% in both the Silk and Silk-K3PO4

coating, suggesting water insolubility [45]. Notably, random coilstransformed into silk I (b-turns) when K3PO4 was added to the silkfibroin.

3.2. Adhesion and mechanical property

Nano-indentation is widely used to measure the surfacemechanical properties [46,47]. The results of the adhesion resis-tance are shown in Fig. 3(A-C). In the loading procedure, the firstplateau of the vertical displacement was at approximately260 nm, which implied that the tip pierced the Silk-KP coating,according to the thickness of the Silk-KP coating (Fig. S1). At thatpoint, the loading time was 19 s, and its corresponding lateral forcewas 820 mN. This lateral force was the indicator of adhesion resis-tance and much higher than that of the other polymer coatings onmagnesium alloys [46], inferring preferable adhesion of the Silk-KPcoating. The hardness, Young’s modulus, and friction coefficient are

Table 1Percent of secondary structures in the Silk and the Silk-K3PO4 coating.

Assignments Silk Silk-K3PO4 coating

Side chains (1590–1605 cm�1) 0.6% ± 0.1 1.0% ± 0.1Silk II, b-sheets (1610–1635 cm�1,

1695–1700 cm�1)32.8% ± 0.2 33.9% ± 0.3

Random coils (1635–1645 cm�1) 15.9% ± 0.3 10.1% ± 0.1Silk I, type II b-turns (1647–1654 cm�1) 15.8% ± 0.3 22.4% ± 0.2a-Helices (1658–1664 cm�1) 13.9% ± 0.2 10.2% ± 0.4Turns and bends (1666–1695 cm�1) 21.0% ± 0.3 22.4% ± 0.2

Fig. 3. (A) Vertical displacement and (B) Lateral force changed with time during the nano-scratch resistance, (C) deformation morphology after nano-scratch, and (D)mechanical property parameters of the Silk-KP (n = 5).




one-way analysis of variance (ANOVA), and the significance wasregarded at p value < 0.05.

3. Results

3.1. Chemical composition of the surface coating

The chemical and structural changes after K3PO4 was added tothe silk fibroin were determined by XPS and ATR-FTIR. As shownin Fig. 2 (A1-A2), the survey spectrum of XPS results showed thatthe main elements on the surfaces of both the Silk-K3PO4 coatingand the Silk were C, N, and O, which were the main elements of silkfibroin. Additionally, the P element was detected on the Silk-K3PO4

coating, which confirmed the addition of K3PO4. Furthermore, theC1s spectrum for the Silk-K3PO4 coating (Fig. 2(A3)) showed char-acteristic peaks at 285 (C–C), 286 (C–H), and 288 (C@O), corre-sponding to different chemical environments of silk fibroin [41,42].

As shown in Fig. 2(B1), the functional groups of both the Silk-K3PO4 coating and the Silk were detected by FTIR, and two peaksat 1655 cm�1 and 1540 cm�1 were found for these two samples,respectively. It was attributed to the amide I (N–H deformation)and amide II (C–N stretching) [43], respectively, which were char-acteristic peaks of silk fibroin. Moreover, the peaks at 1065 cm�1,982 cm�1, and 824 cm�1 were assigned to v3 vibration peak ofPO4

3� [44], which was obtained for the Silk-K3PO4 coating only.To further realize the secondary structural changes of silk fibroinafter K3PO4 was added to the silk fibroin, the amide I region(1590–1700 cm�1) was further characterized by deconvolution ofthe ATR-FTIR spectra. The proportion of the main secondary struc-tures was calculated, and the data are given in Table 1. The contentof b-sheets was more than 30% in both the Silk and Silk-K3PO4

coating, suggesting water insolubility [45]. Notably, random coilstransformed into silk I (b-turns) when K3PO4 was added to the silkfibroin.

3.2. Adhesion and mechanical property

Nano-indentation is widely used to measure the surfacemechanical properties [46,47]. The results of the adhesion resis-tance are shown in Fig. 3(A-C). In the loading procedure, the firstplateau of the vertical displacement was at approximately260 nm, which implied that the tip pierced the Silk-KP coating,according to the thickness of the Silk-KP coating (Fig. S1). At thatpoint, the loading time was 19 s, and its corresponding lateral forcewas 820 mN. This lateral force was the indicator of adhesion resis-tance and much higher than that of the other polymer coatings onmagnesium alloys [46], inferring preferable adhesion of the Silk-KPcoating. The hardness, Young’s modulus, and friction coefficient are

Table 1Percent of secondary structures in the Silk and the Silk-K3PO4 coating.

Assignments Silk Silk-K3PO4 coating

Side chains (1590–1605 cm�1) 0.6% ± 0.1 1.0% ± 0.1Silk II, b-sheets (1610–1635 cm�1,

1695–1700 cm�1)32.8% ± 0.2 33.9% ± 0.3

Random coils (1635–1645 cm�1) 15.9% ± 0.3 10.1% ± 0.1Silk I, type II b-turns (1647–1654 cm�1) 15.8% ± 0.3 22.4% ± 0.2a-Helices (1658–1664 cm�1) 13.9% ± 0.2 10.2% ± 0.4Turns and bends (1666–1695 cm�1) 21.0% ± 0.3 22.4% ± 0.2

Fig. 3. (A) Vertical displacement and (B) Lateral force changed with time during the nano-scratch resistance, (C) deformation morphology after nano-scratch, and (D)mechanical property parameters of the Silk-KP (n = 5).



shown in Fig. 3(D). According to the evaluation of other silk fibroincoatings [48], the Silk-KP was found to be robust because of thehigher hardness, higher Young’s modulus, and lower frictioncoefficient.

3.3. Corrosion resistance in vitro

Corrosion of magnesium alloys in the physiological environ-ment was accompanied by hydrogen release and pH valueincrease. The corrosion process occurred as per the followingequation:

Mgþ 2H2O ! Mg2þ þ 2OH� þH2 "Thus, the hydrogen release volume and pH value changes

reflected the corrosion status of each group during 14-day immer-sion in Hank’s solution. As shown in Fig. 4(A, B), the hydrogenrelease of the Silk-KP was an average of 1.19 ml/cm2, and the pHvalue was increased to 8.73 at 14 days. However, the bare Mg-1Ca alloy showed a quick and copious accumulation of hydrogen

Fig. 4. Variation of (A) hydrogen evolution volume and (B) pH values, (C) surface morphologies of each group immersed in Hank’s solution for 7 days, (D) ATR-FTIR spectraand XPS survey spectra of samples after immersion in Hank’s solution for 7 days, and (E) XPS co-responding core-level spectra for C 1s and P 2p of the Silk-KP after 7 days ofimmersion in Hank’s solution (n = 3).

Table 3Changes in the length, width, and depth of the scratches on the surface of the Silk andSilk-KP immersion in Hank’s solution for 24 h; the ‘‘+” represents increase, while the‘‘�” represents decrease.

Mg-1Ca alloy Silk Silk-KP

D length (lm) +0.15 �0.30 �1.47D width (lm) �0.40 +0.07 �0.47D depth (lm) +0.27 +0.14 �0.87

Table 2EDX results of each group immersion in Hank’s solution for 7 days.

At% In Hank’s solution

EL. Mg-1Ca Alloy Silk Silk-KP

C K – 8 12O K 59 21 30Mg K 11 25 15P K 9 14 26Ca K 22 9 16F k – 22 1




evolved with time (5.75 ml/cm2 at 14 days), and the pH value washigher than 9 in the first three days and reached to 10.13 at14 days.

The surface morphologies of experimental samples after 7-dayimmersion in Hank’s solution were studied by SEM, as shown inFig. 4(C). The Silk and the Silk-KP groups showed uniform surfacemorphologies with a few corrosion products on the surface. Afterremoving from Hank’s solution and air-drying, cracks appeared;the cracks on the Silk group were wider than those on the Silk-KP group. On the other hand, the bare Mg-1Ca alloy was coveredwith thick corrosion products.

The surface elements (orange block region framed in Fig. 4(C))were detected by EDX, and the results are given in Table 2. The

surface of the Silk-KP had more P and C elements but fewer Mgand F elements. Furthermore, the corrosion products were ana-lyzed by ATR-FTIR and XPS. As displayed in Fig. 4(D), the character-istic peaks of CO3

3�, PO43�, and HPO4

2� appeared in the infraredspectra, implying mainly corrosion products of carbonate, phos-phate, and hydrophosphate. Notably, the characteristic peaks ofamide were detected (marked in red circle) in the Silk-KP group.XPS results revealed that the main elements on the surface wereMg, O, P, C, and N, which only appeared on the surface of the Silkand Silk-KP, thereby confirming the existence of the silk fibroincoating. The C1s spectrum of the Silk-KP showed characteristicpeaks of silk fibroin (C–C, C–H, and C@O) and carbonate. The P2p spectrum was a single peak, implying PO4

3�. Overall, Silk-PK

Fig. 5. Optical imaging and EDX spectra results of scratch tests of the bare Mg-1Ca alloy, Silk, and the Silk-KP.




evolved with time (5.75 ml/cm2 at 14 days), and the pH value washigher than 9 in the first three days and reached to 10.13 at14 days.

The surface morphologies of experimental samples after 7-dayimmersion in Hank’s solution were studied by SEM, as shown inFig. 4(C). The Silk and the Silk-KP groups showed uniform surfacemorphologies with a few corrosion products on the surface. Afterremoving from Hank’s solution and air-drying, cracks appeared;the cracks on the Silk group were wider than those on the Silk-KP group. On the other hand, the bare Mg-1Ca alloy was coveredwith thick corrosion products.

The surface elements (orange block region framed in Fig. 4(C))were detected by EDX, and the results are given in Table 2. The

surface of the Silk-KP had more P and C elements but fewer Mgand F elements. Furthermore, the corrosion products were ana-lyzed by ATR-FTIR and XPS. As displayed in Fig. 4(D), the character-istic peaks of CO3

3�, PO43�, and HPO4

2� appeared in the infraredspectra, implying mainly corrosion products of carbonate, phos-phate, and hydrophosphate. Notably, the characteristic peaks ofamide were detected (marked in red circle) in the Silk-KP group.XPS results revealed that the main elements on the surface wereMg, O, P, C, and N, which only appeared on the surface of the Silkand Silk-KP, thereby confirming the existence of the silk fibroincoating. The C1s spectrum of the Silk-KP showed characteristicpeaks of silk fibroin (C–C, C–H, and C@O) and carbonate. The P2p spectrum was a single peak, implying PO4

3�. Overall, Silk-PK

Fig. 5. Optical imaging and EDX spectra results of scratch tests of the bare Mg-1Ca alloy, Silk, and the Silk-KP.



showed higher intensity of silk fibroin characteristic elements andPO4

3�, signifying more stability of the Silk-KP and more phosphatedeposition.

3.4. Self-healing capacity

Artificial scratch was made on the bare Mg-1Ca alloy, the Silk,and the Silk-KP samples using a Nanoindenter System. The depth,width, and length of the resulting scratch were measured and com-pared with values before and after immersion in Hank’s solutionfor 24 h. The changes occurring in the 3-dimensions were analyzedusing supporting software, which are given in Table 3. As shown inthe images (Fig. 5), the scratch on the Silk-KP sample was shorterand narrower. The measurement of 3-dimensional size also con-firmed the healing of the crack on the Silk-KP. By EDX analysis atthe position of scratches after immersion, it was revealed thescratch region of the Silk-KP was covered with more phosphatethan the top portion of the Silk because of the higher P elementconcentration of the Silk-KP than that of the Silk.

Scanning vibrating electrode technique (SVET), which serves asa nondestructive tool to visualize current densities and accuratelyexpress the localized cathodic and anodic activities, is employed tofurther evaluate the self-healing ability of the experimental sam-ples in the electrolyte solution. The artificial dot defects wereapproximately 200 ± 5 lm in diameter and depth. While immer-sion in Hank’s solution, the evolution of ionic flux signals aroundartificial defects was recorded and transformed to current densitydata, which are presented in Fig. 6 as 3D maps. For the Silk-KPgroup, after immersion for 40 min, the anodic current peakappeared, thus implying that the corrosion occurred around thedefects. While immersion for 80 min, the anodic current peakincreased, illustrating progressive propagation of localized corro-sion. After immersion for 120 min, both the anodic and cathodic

current density verged to zero plane (Fig. 6 (B3)), verifying thatthe designed self-healing function operated normally. The mapsof the Silk-KP group were similar to those given in the previousreports of self-healing coatings on metals [49]. However, duringimmersion, the Silk group sustained a negative peak evolution,representing significant cathodic current density. It was illustratedthat the material chose anodic sacrifice protection to prevent far-ther corrosion [50]. Furthermore, when immersion for 160 min,the Silk demonstrated a deeper cathodic current peak, signifyingmore serious corrosion on the defect region.

The EIS measurement was a nondestructive and powerfultechnique for the assessment of corrosion resistance properties.The Nyquist plots and Bode magnitude plots of the Silk andSilk-KP groups in Hank’s solution with different immersion timepoints are presented in Fig. 7 (A for the Silk group, and B forthe Silk-KP group). To further interpret the EIS results, the equiv-alent electrical circuit (EEC) models (Fig. 7 (A4) and (B4)) wereproposed for fitting the data. Constant phase element (CPE) wasused in the fitting procedure, and Rs was the solution resistance.The EEC model for the Silk-KP group included two parametersand expressed as Rs(Qdl(QbRb)Rct). The Silk group displayed onemore inductive semicircle arc at low frequency, and the EC wasdenoted as Rs(QdlRct)(QbRb(RLL)). Based on the circuits, thedouble-layer capacitance (Qdl) and Rct can be determined at highfrequency. The capacitive (Qb) and resistance (Rb) were indicativeof the coatings on the surface of Mg-1Ca alloy. The pseudo-inductive loop at low frequency appearing in the Silk was relatedto two parameters, RL and L, which implied that the dissolutionand pitting corrosion occurred during immersion [51]. Moreover,polarization resistance (Rp) was defined as the differencebetween the real impedance of the Nyquist plot. While the fre-quency was equal to zero with solution resistance (Rs), Rpwas proportional to the corrosion resistance [52] and could be

Fig. 6. SVET 3D maps of the electric current density measured above the defected surface of the Silk (A1-A4) and the Silk-KP (B1-B4).




Fig. 7. EIS and the fitted results for the Silk (A1-A4) and the Silk-KP (B1-B4) (1–4: Nyquist plots, Bode plots of |Z| vs. frequency, Bode plots of phase angle vs. frequency,respectively, and equivalent circuit model immersion in Hank’s solution at various time points.




Fig. 7. EIS and the fitted results for the Silk (A1-A4) and the Silk-KP (B1-B4) (1–4: Nyquist plots, Bode plots of |Z| vs. frequency, Bode plots of phase angle vs. frequency,respectively, and equivalent circuit model immersion in Hank’s solution at various time points.



calculated with the following equations for the correspondingmodels of the Silk and Silk-KP groups immersed for various timepoints

Fig. 7 (A4):

1RP

¼ 1

Rctþ R-1b þR-1

L

� �-1 ð1Þ

Fig. 7 (B4):

1RP

¼ 1RctþRb

ð2Þ

Rp values are listed in Table 4. It was found that the Rp of theSilk group decreased with prolonged immersion time, but the Rp

of the Silk-KP group just decreased within immersion for 40–80 min and increased later.

3.5. pH Stimuli-Responsive ability

The release profiles of PO43� are displayed in Fig. 8. While

immersion for only 30 min, the Silk-KP immersed in the DI waterwith a pH of 10.0 released much more PO4

3� (55 ± 2 mg/cm2,n = 5) than that immersed in the neutral solution (18 ± 2 mg/cm2,n = 5). During the first 6 h, the cumulative release of PO4

3� atpH = 10.0 (260 ± 20 mg/cm2, n = 5) was more than two times of thatat pH = 7.4 (107 ± 12 mg/cm2, n = 5). It was observed that the Silk-KP more rapidly released the PO4

3� in an alkaline environment,which was deemed as pH stimuli-responsiveness.

3.6. In vitro biocompatibility and osteogenic activity

3.6.1. Cell attachmentCell attachment was investigated through F-actin/FAK staining

(indirect contact culture) and SEM observation (direct culture),respectively. As displayed in Fig. 9(A) and (B), more cells and actinfilaments were found in cells cultured in the Silk and Silk-KPextracts. Additionally, cells of the Silk and Silk-KP groups exhibitedfusiform-spreading morphologies with elongated filopodia (whitearrows in Fig. 9B), presenting tight adhesion to the substrates. Asfor the bare Mg-1Ca alloy, especially for cells directly cultured onthe substrate, a few near-round dead cells and cracks weredetected on the surface.

3.6.2. Cell cytotoxicityThe CCK-8 cell viability tests were used to evaluate the cytotox-

icity of each group for cells being cultured for 1, 3, and 5 days(Fig. 9(C)). The results were expressed as average activity withstandard deviation. At 1 and 3 days, the cell viability of each groupshowed no significant difference. At 5 days, the cell viability of theSilk-KP group reached 114% of the negative control, revealing thatthe Silk-KP group was noncytotoxic, and when prolonging the cul-ture time, it might promote cell proliferation.

3.6.3. Osteo-differentiationALP activity, as an early hallmark for osteogenic differentiation

potential, was stained with BCIP/NBT, as shown in Fig. 10(A), andwas normalized and calculated, as shown in Fig. 10(B). The ALPactivity for all groups increased from 7 days to 14 days. While cul-tured for 7 days, the ALP activities of all groups showed no statis-

Table 4Fitting results of EIS data for the Silk and the Silk-KP immersion in Hank’s solution at various time points.

Sample Time (min) Rs (X cm2) Rct (kX cm2) Qdl (mFsn�1 cm�2) n1 Rb (kX cm2) Qb (mFsn�1 cm�2) n2 RL (X cm2) L (kH cm�2) Rp (kX cm2)

Silk 40 65.02 56.70 4.49 0.86 10.82 205.77 0.49 26.69 283.67 6480 65.04 45.70 4.57 0.86 9.02 228.66 0.50 28.65 298.96 52120 69.76 31.39 5.51 0.91 6.26 16.77 0.73 1091.15 1786.86 37160 60.81 19.00 10.34 0.77 4.83 9.38 0.92 3563.90 1748.13 24

Silk-KP 40 140.09 88.44 5.34 0.82 6.77 5.45 0.72 – – 9580 142.42 65.45 5.14 0.83 5.00 5.55 0.79 – – 71120 134.74 160.99 5.21 0.81 14.96 5.22 0.68 – – 166160 133.36 170.57 5.28 0.81 14.23 5.64 0.32 – – 176

Fig. 8. The release behavior of PO43�: (A) release amount at different time points, (B) cumulative release amount of the Silk-KP-coated glass immersion in DI water with

varying pH values for 6 and12 h (n = 5, **p < 0.005, ***p < 0.001).




tical difference. Nevertheless, the Silk-KP group showed a higherALP activity expression while cultured for 14 days. Additionally,the staining results coincided well with the ALP activity results.The Silk-KP group displayed deeper staining feature. The ECM col-lagen was secreted at a similar level among all experimentalgroups. Moreover, as for ARS staining of calcium, the Silk andSilk-KP groups demonstrated more locally better stained spots/nodules and higher absorbance in quantitative analysis. It wasrevealed that both the Silk and Silk-KP groups could facilitatematrix mineralization.

4. Discussion

Drawing inspiration from the self-healing coatings on magne-sium alloys in the industrial field, a biocompatible self-healingcoating on the biomedical magnesium alloys was developed in thiswork. Compositing the silk fibroin and K3PO4 inorganic salt to formthe self-healing coating could combine loading capacity, biocom-patibility, and pH sensitivity from silk fibroin and the insolubilityand osteogenesis from phosphate. Hence, it was promising to serveas a self-healing anticorrosion coating on biodegradable magne-sium alloys, especially for use within the bone.

The coating composition characteristics confirmed that thedesigned coating could be coated on the magnesium alloysubstrate successfully. Moreover, the structures of the resultingcoatings kept stable for a certain period, which was a crucial prop-erty for anticorrosion coating on magnesium alloys. The fabricationmethod of spin-coating and the addition of K+ ions assisted inadjusting the secondary structures of silk fibroin. Previously, thespin-coating method provided the shear force to regulate the con-tent of b-sheets higher than 30% [30]. In other works, a silk fibroinfilm fabricated by casting was soluble (content of b-sheets below25%) [45]. Notably, random coils transformed into silk I (b-turns),while K3PO4 was encapsulated in silk fibroin. A previous workshowed that the K+ ion regulates the secondary structure of silkfibroin [53]. It was suggested that K+ ions, at a high concentrationin the silk fibroin solution (higher than 3.7 mg/g), transform ran-dom coils to silk I because of the interaction of the K+ ions with car-bonyl oxygen atoms from amino acid palindrome sequences(illustrated in Fig. 2(C2)). Moreover, the comprehensive mechanicalperformance tests displayed that the Silk-KP was robust and had apreferable adhesion to the substrate. These properties are benefi-cial for the coatings of bone implants. The higher Young’s modulusand preferable adhesion make it possible to be coated on thesurface of a variety of shaped substrates, such as screws, plate,

Fig. 9. (A) Actin-nucleus counter-staining and (B) SEM morphologies of cells for each group cultured at 6 and 24 h; (C) Viability of cells cultured in the extracts for 1, 3, and5 days by CCK-8 assay kit results (n = 5, ***p < 0.001).




tical difference. Nevertheless, the Silk-KP group showed a higherALP activity expression while cultured for 14 days. Additionally,the staining results coincided well with the ALP activity results.The Silk-KP group displayed deeper staining feature. The ECM col-lagen was secreted at a similar level among all experimentalgroups. Moreover, as for ARS staining of calcium, the Silk andSilk-KP groups demonstrated more locally better stained spots/nodules and higher absorbance in quantitative analysis. It wasrevealed that both the Silk and Silk-KP groups could facilitatematrix mineralization.

4. Discussion

Drawing inspiration from the self-healing coatings on magne-sium alloys in the industrial field, a biocompatible self-healingcoating on the biomedical magnesium alloys was developed in thiswork. Compositing the silk fibroin and K3PO4 inorganic salt to formthe self-healing coating could combine loading capacity, biocom-patibility, and pH sensitivity from silk fibroin and the insolubilityand osteogenesis from phosphate. Hence, it was promising to serveas a self-healing anticorrosion coating on biodegradable magne-sium alloys, especially for use within the bone.

The coating composition characteristics confirmed that thedesigned coating could be coated on the magnesium alloysubstrate successfully. Moreover, the structures of the resultingcoatings kept stable for a certain period, which was a crucial prop-erty for anticorrosion coating on magnesium alloys. The fabricationmethod of spin-coating and the addition of K+ ions assisted inadjusting the secondary structures of silk fibroin. Previously, thespin-coating method provided the shear force to regulate the con-tent of b-sheets higher than 30% [30]. In other works, a silk fibroinfilm fabricated by casting was soluble (content of b-sheets below25%) [45]. Notably, random coils transformed into silk I (b-turns),while K3PO4 was encapsulated in silk fibroin. A previous workshowed that the K+ ion regulates the secondary structure of silkfibroin [53]. It was suggested that K+ ions, at a high concentrationin the silk fibroin solution (higher than 3.7 mg/g), transform ran-dom coils to silk I because of the interaction of the K+ ions with car-bonyl oxygen atoms from amino acid palindrome sequences(illustrated in Fig. 2(C2)). Moreover, the comprehensive mechanicalperformance tests displayed that the Silk-KP was robust and had apreferable adhesion to the substrate. These properties are benefi-cial for the coatings of bone implants. The higher Young’s modulusand preferable adhesion make it possible to be coated on thesurface of a variety of shaped substrates, such as screws, plate,

Fig. 9. (A) Actin-nucleus counter-staining and (B) SEM morphologies of cells for each group cultured at 6 and 24 h; (C) Viability of cells cultured in the extracts for 1, 3, and5 days by CCK-8 assay kit results (n = 5, ***p < 0.001).



Fig. 10. Osteogenic activity of MC3T3 cultured in different extracts. (A) Optical imaging of the coloration of ALP activity (deep purple) for 7 and 14 days, collagen (COL)secretion (purplish red) for 28 days, and calcium (CAL) deposition (red or purplish red) for 28 days; (B) the degree of ALP activity for 7 and 14 days; (C) quantification ofcollagen secretion and calcium deposition for 28 days (n = 5, ***p < 0.001). (For interpretation of the references to color in this figure legend, the reader is referred to the webversion of this article.)

Fig. 11. Schematic of the self-healing mechanism upon Silk-KP exposure in the immersion environment.




needles, etc. The hardness and wear resistance can avoid physicaldamage to the coating during the implantation, to some extent.

According to the immersion testing results in vitro, the Silk-KPshowed preferable corrosion resistance as compared to the bareMg-1Ca alloy. Although the cracks appeared on the Silk-KP afterimmersion for 7 days, which might be due to air-drying beforeSEM observation, the increase in pH value could be kept at a lim-ited range and low hydrogen release volume. It was observed thatthe coatings could avoid the foremost issues presenting in biomed-ical magnesium alloys [54]. Moreover, the cracks appearing on thecoatings were common for polymer coatings on the magnesiumalloys during immersion [55].

Self-healing capacity was one of the main parameters focused inthis work, and the self-healing process could be speculated as illus-trated in Fig. 11. Once the magnesium alloy was corroded, defectsappeared on the surface with increase in local pH value and releaseof Mg2+ ions (diagramed in Fig. 11(B)), presenting a decrease incorrosion resistance (expression in the increase of anodic currentdensity in SVET and decrease of Rp in EIS results). As described insection 3.5, the Silk-KP could release the PO4

3� ions rapidly in analkaline environment. This was attributed to the residual randomcoils and a-helical structures in the Silk-KP, which endowed thestructure reversibility of the Silk-KP. While in a microenvironmentwith high pH, these structures of the silk fibroin were of negativecharge and the interactions were disconnected, resulting in a likelyrather elongated molecular conformation [56]. Thus, more PO4

3�

ions were released in the local defects (illustrated at the bottomof Fig. 11). The released Mg2+ ions would react with PO4

3� ions toyield the product Mg3(PO4)2, as per the equation:

3Mg2þ þ 2PO3�4 ! Mgð Þ3 PO4ð Þ2 #

Generally, PO43� ions and the Mg3(PO4)2 salt are stable in an

alkaline environment, and the solubility product constant ofMg3(PO4)2 is as low as 1.04 * 10�24. Thus, the Mg3(PO4)2 salt coulddeposit a passive film on the defects, healing the barrier propertyand preventing further corrosion (Fig. 11(C)). This was demon-strated by the diminished scratch size on the surface, the currentdensities verged to zero plane in SVET and increased Rp of the EIS.

The cytocompatibility and osteogenic activities were evaluatedby the multiple cell responses in vitro. The causes for the improve-ment of biocompatibility of the Silk-KP could be attributed to thefollowing factors: (1) the improvement in corrosion resistance.Rapid corrosion is accompanied with locally unduly high pH valuesand Mg2+ concentration, which could destroy the protein-basedtransport channels and then disabling active ionic transport [57].(2) Biocompatibility and osteogenic activity of the silk fibroin coat-ing itself. Previous works confirmed the silk fibroin would enhancethe expression of ALP and Runx2 mRNA by significantly downreg-ulating Notch signaling [58]. (3) Addition of phosphate. Phosphatewas the inorganic ingredient of natural bone. A previous studyindicated good osteoinduction ability of phosphate [59].

5. Conclusions

The biocompatible pH-sensitive self-healing anti-corrosive Silk-KPwas fabricated on theMg-1Ca alloy by a convenientmethod, andit yielded the following advantages over the bare Mg-1Ca alloy.

(1) Corrosion resistance was enhanced by the self-healing func-tion of Silk-KP, and Silk-KP showed more stable structure bycompositing silk fibroin and K+ ions.

(2) PO43� ions existed in Silk-KP, serving as the corrosion inhibi-

tor, and the scratch tests, SVET, and EIS were performed toconfirm the self-healing ability of the Silk-KP coating ontop of the bare Mg-1Ca alloy.

(3) Biocompatibility evaluation in vitro revealed the designedSilk-KP not only was biocompatible but also exhibited osteo-genic activity.

This work introduces self-healing coating on biodegradablemagnesium alloys and provides a new candidate coating for sur-face modification of biodegradable metals for potential use withinthe bone.

Acknowledgments

This work was supported by the National Key Research andDevelopment Program of China (Grant No. 2018YFC1106600),National Natural Science Foundation of China (Grant Nos.51431002, 31670974, and 51871004), NSFC/RGC Joint ResearchScheme (Grant No. 5161101031), and Beijing Municipal HealthCommision (Grant No. BMC2018-4).

Appendix A. Supplementary data

Supplementary data to this article can be found online athttps://doi.org/10.1016/j.actbio.2019.04.045.

References

[1] D. Zhao, F. Witte, F. Lu, J. Wang, J. Li, L. Qin, Current status on clinicalapplications of magnesium-based orthopaedic implants: A review from clinicaltranslational perspective, Biomaterials 112 (2017) 287–302.

[2] Y. Chen, Z. Xu, C. Smith, J. Sankar, Recent advances on the development ofmagnesium alloys for biodegradable implants, Acta biomaterialia 10 (11)(2014) 4561–4573.

[3] F. Witte, J. Fischer, J. Nellesen, H.A. Crostack, V. Kaese, A. Pisch, F. Beckmann, H.Windhagen, In vitro and in vivo corrosion measurements of magnesium alloys,Biomaterials 27 (7) (2006) 1013–1018.

[4] F. Witte, V. Kaese, H. Haferkamp, E. Switzer, A. Meyer-Lindenberg, C.J. Wirth, H.Windhagen, In vivo corrosion of four magnesium alloys and the associatedbone response, Biomaterials 26 (17) (2005) 3557–3563.

[5] G. Song, Control of biodegradation of biocompatable magnesium alloys,Corrosion Science 49 (4) (2007) 1696–1701.

[6] J. Tang, J. Wang, X. Xie, P. Zhang, Y. Lai, Y. Li, L. Qin, Surface coating reducesdegradation rate of magnesium alloy developed for orthopaedic applications,Journal of Orthopaedic Translation 1 (1) (2013) 41–48.

[7] J. Han, P. Wan, Y. Sun, Z. Liu, X. Fan, L. Tan, K. Yang, Fabrication and Evaluationof a Bioactive Sr–Ca–P Contained Micro-Arc Oxidation Coating on MagnesiumStrontium Alloy for Bone Repair Application, Journal of Materials Science &Technology 32 (3) (2016) 233–244.

[8] L. Zhao, C. Cui, X. Wang, S. Liu, S. Bu, Q. Wang, Y. Qi, Corrosion resistance andcalcium–phosphorus precipitation of micro-arc oxidized magnesium forbiomedical applications, Applied Surface Science 330 (2015) 431–438.

[9] Z. Yao, Q. Xia, L. Chang, C. Li, Z. Jiang, Structure and properties of compoundcoatings on Mg alloys by micro-arc oxidation/hydrothermal treatment, Journalof Alloys and Compounds 633 (2015) 435–442.

[10] Z.J. Jia, M. Li, Q. Liu, X.C. Xu, Y. Cheng, Y.F. Zheng, T.F. Xi, S.C. Wei, Micro-arcoxidization of a novel Mg–1Ca alloy in three alkaline KF electrolytes: Corrosionresistance and cytotoxicity, Applied Surface Science 292 (2014) 1030–1039.

[11] M.D. Pereda, C. Alonso, L. Burgos-Asperilla, J.A. del Valle, O.A. Ruano, P. Perez,M.A. Fernandez, Lorenzo de Mele, Corrosion inhibition of powder metallurgyMg by fluoride treatments, Acta biomaterialia 6 (5) (2010) 1772–1782.

[12] L. Mao, G. Yuan, J. Niu, Y. Zong, W. Ding, In vitro degradation behavior andbiocompatibility of Mg-Nd-Zn-Zr alloy by hydrofluoric acid treatment,Materials science & engineering, C, Materials for biological applications 33(1) (2013) 242–250.

[13] J. Sun, Y. Zhu, L. Meng, W. Wei, Y. Li, X. Liu, Y. Zheng, Controlled release andcorrosion protection by self-assembled colloidal particles electrodepositedonto magnesium alloys, Journal of Materials Chemistry B 3 (8) (2015) 1667–1676.

[14] P. Xiong, Z. Jia, M. Li, W. Zhou, J. Yan, Y. Wu, Y. Cheng, Y. Zheng, Biomimetic Ca,Sr/P-Doped Silk Fibroin Films on Mg-1Ca Alloy with Dramatic CorrosionResistance and Osteogenic Activities, ACS Biomaterials Science & Engineering4 (9) (2018) 3163–3176.

[15] X. Ma, C. Blawert, D. Höche, K.U. Kainer, M.L. Zheludkevich, Simulation assistedinvestigation of substrate geometry impact on PEO coating formation, Surfaceand Coatings Technology 350 (2018) 281–297.

[16] S.B. Ulaeto, R. Rajan, J.K. Pancrecious, T.P.D. Rajan, B.C. Pai, Developments insmart anticorrosive coatings with multifunctional characteristics, Progress inOrganic Coatings 111 (2017) 294–314.




needles, etc. The hardness and wear resistance can avoid physicaldamage to the coating during the implantation, to some extent.

According to the immersion testing results in vitro, the Silk-KPshowed preferable corrosion resistance as compared to the bareMg-1Ca alloy. Although the cracks appeared on the Silk-KP afterimmersion for 7 days, which might be due to air-drying beforeSEM observation, the increase in pH value could be kept at a lim-ited range and low hydrogen release volume. It was observed thatthe coatings could avoid the foremost issues presenting in biomed-ical magnesium alloys [54]. Moreover, the cracks appearing on thecoatings were common for polymer coatings on the magnesiumalloys during immersion [55].

Self-healing capacity was one of the main parameters focused inthis work, and the self-healing process could be speculated as illus-trated in Fig. 11. Once the magnesium alloy was corroded, defectsappeared on the surface with increase in local pH value and releaseof Mg2+ ions (diagramed in Fig. 11(B)), presenting a decrease incorrosion resistance (expression in the increase of anodic currentdensity in SVET and decrease of Rp in EIS results). As described insection 3.5, the Silk-KP could release the PO4

3� ions rapidly in analkaline environment. This was attributed to the residual randomcoils and a-helical structures in the Silk-KP, which endowed thestructure reversibility of the Silk-KP. While in a microenvironmentwith high pH, these structures of the silk fibroin were of negativecharge and the interactions were disconnected, resulting in a likelyrather elongated molecular conformation [56]. Thus, more PO4

3�

ions were released in the local defects (illustrated at the bottomof Fig. 11). The released Mg2+ ions would react with PO4

3� ions toyield the product Mg3(PO4)2, as per the equation:

3Mg2þ þ 2PO3�4 ! Mgð Þ3 PO4ð Þ2 #

Generally, PO43� ions and the Mg3(PO4)2 salt are stable in an

alkaline environment, and the solubility product constant ofMg3(PO4)2 is as low as 1.04 * 10�24. Thus, the Mg3(PO4)2 salt coulddeposit a passive film on the defects, healing the barrier propertyand preventing further corrosion (Fig. 11(C)). This was demon-strated by the diminished scratch size on the surface, the currentdensities verged to zero plane in SVET and increased Rp of the EIS.

The cytocompatibility and osteogenic activities were evaluatedby the multiple cell responses in vitro. The causes for the improve-ment of biocompatibility of the Silk-KP could be attributed to thefollowing factors: (1) the improvement in corrosion resistance.Rapid corrosion is accompanied with locally unduly high pH valuesand Mg2+ concentration, which could destroy the protein-basedtransport channels and then disabling active ionic transport [57].(2) Biocompatibility and osteogenic activity of the silk fibroin coat-ing itself. Previous works confirmed the silk fibroin would enhancethe expression of ALP and Runx2 mRNA by significantly downreg-ulating Notch signaling [58]. (3) Addition of phosphate. Phosphatewas the inorganic ingredient of natural bone. A previous studyindicated good osteoinduction ability of phosphate [59].

5. Conclusions

The biocompatible pH-sensitive self-healing anti-corrosive Silk-KPwas fabricated on theMg-1Ca alloy by a convenientmethod, andit yielded the following advantages over the bare Mg-1Ca alloy.

(1) Corrosion resistance was enhanced by the self-healing func-tion of Silk-KP, and Silk-KP showed more stable structure bycompositing silk fibroin and K+ ions.

(2) PO43� ions existed in Silk-KP, serving as the corrosion inhibi-

tor, and the scratch tests, SVET, and EIS were performed toconfirm the self-healing ability of the Silk-KP coating ontop of the bare Mg-1Ca alloy.

(3) Biocompatibility evaluation in vitro revealed the designedSilk-KP not only was biocompatible but also exhibited osteo-genic activity.

This work introduces self-healing coating on biodegradablemagnesium alloys and provides a new candidate coating for sur-face modification of biodegradable metals for potential use withinthe bone.

Acknowledgments

This work was supported by the National Key Research andDevelopment Program of China (Grant No. 2018YFC1106600),National Natural Science Foundation of China (Grant Nos.51431002, 31670974, and 51871004), NSFC/RGC Joint ResearchScheme (Grant No. 5161101031), and Beijing Municipal HealthCommision (Grant No. BMC2018-4).

Appendix A. Supplementary data

Supplementary data to this article can be found online athttps://doi.org/10.1016/j.actbio.2019.04.045.

References

[1] D. Zhao, F. Witte, F. Lu, J. Wang, J. Li, L. Qin, Current status on clinicalapplications of magnesium-based orthopaedic implants: A review from clinicaltranslational perspective, Biomaterials 112 (2017) 287–302.

[2] Y. Chen, Z. Xu, C. Smith, J. Sankar, Recent advances on the development ofmagnesium alloys for biodegradable implants, Acta biomaterialia 10 (11)(2014) 4561–4573.

[3] F. Witte, J. Fischer, J. Nellesen, H.A. Crostack, V. Kaese, A. Pisch, F. Beckmann, H.Windhagen, In vitro and in vivo corrosion measurements of magnesium alloys,Biomaterials 27 (7) (2006) 1013–1018.

[4] F. Witte, V. Kaese, H. Haferkamp, E. Switzer, A. Meyer-Lindenberg, C.J. Wirth, H.Windhagen, In vivo corrosion of four magnesium alloys and the associatedbone response, Biomaterials 26 (17) (2005) 3557–3563.

[5] G. Song, Control of biodegradation of biocompatable magnesium alloys,Corrosion Science 49 (4) (2007) 1696–1701.

[6] J. Tang, J. Wang, X. Xie, P. Zhang, Y. Lai, Y. Li, L. Qin, Surface coating reducesdegradation rate of magnesium alloy developed for orthopaedic applications,Journal of Orthopaedic Translation 1 (1) (2013) 41–48.

[7] J. Han, P. Wan, Y. Sun, Z. Liu, X. Fan, L. Tan, K. Yang, Fabrication and Evaluationof a Bioactive Sr–Ca–P Contained Micro-Arc Oxidation Coating on MagnesiumStrontium Alloy for Bone Repair Application, Journal of Materials Science &Technology 32 (3) (2016) 233–244.

[8] L. Zhao, C. Cui, X. Wang, S. Liu, S. Bu, Q. Wang, Y. Qi, Corrosion resistance andcalcium–phosphorus precipitation of micro-arc oxidized magnesium forbiomedical applications, Applied Surface Science 330 (2015) 431–438.

[9] Z. Yao, Q. Xia, L. Chang, C. Li, Z. Jiang, Structure and properties of compoundcoatings on Mg alloys by micro-arc oxidation/hydrothermal treatment, Journalof Alloys and Compounds 633 (2015) 435–442.

[10] Z.J. Jia, M. Li, Q. Liu, X.C. Xu, Y. Cheng, Y.F. Zheng, T.F. Xi, S.C. Wei, Micro-arcoxidization of a novel Mg–1Ca alloy in three alkaline KF electrolytes: Corrosionresistance and cytotoxicity, Applied Surface Science 292 (2014) 1030–1039.

[11] M.D. Pereda, C. Alonso, L. Burgos-Asperilla, J.A. del Valle, O.A. Ruano, P. Perez,M.A. Fernandez, Lorenzo de Mele, Corrosion inhibition of powder metallurgyMg by fluoride treatments, Acta biomaterialia 6 (5) (2010) 1772–1782.

[12] L. Mao, G. Yuan, J. Niu, Y. Zong, W. Ding, In vitro degradation behavior andbiocompatibility of Mg-Nd-Zn-Zr alloy by hydrofluoric acid treatment,Materials science & engineering, C, Materials for biological applications 33(1) (2013) 242–250.

[13] J. Sun, Y. Zhu, L. Meng, W. Wei, Y. Li, X. Liu, Y. Zheng, Controlled release andcorrosion protection by self-assembled colloidal particles electrodepositedonto magnesium alloys, Journal of Materials Chemistry B 3 (8) (2015) 1667–1676.

[14] P. Xiong, Z. Jia, M. Li, W. Zhou, J. Yan, Y. Wu, Y. Cheng, Y. Zheng, Biomimetic Ca,Sr/P-Doped Silk Fibroin Films on Mg-1Ca Alloy with Dramatic CorrosionResistance and Osteogenic Activities, ACS Biomaterials Science & Engineering4 (9) (2018) 3163–3176.

[15] X. Ma, C. Blawert, D. Höche, K.U. Kainer, M.L. Zheludkevich, Simulation assistedinvestigation of substrate geometry impact on PEO coating formation, Surfaceand Coatings Technology 350 (2018) 281–297.

[16] S.B. Ulaeto, R. Rajan, J.K. Pancrecious, T.P.D. Rajan, B.C. Pai, Developments insmart anticorrosive coatings with multifunctional characteristics, Progress inOrganic Coatings 111 (2017) 294–314.



[17] D.V. Andreeva, D. Fix, H. Mohwald, D.G. Shchukin, Self-Healing AnticorrosionCoatings Based on pH-Sensitive Polyelectrolyte/Inhibitor SandwichlikeNanostructures, Adv Mater 20 (14) (2008) 2789–2794.

[18] Z. Zheng, X. Huang, M. Schenderlein, D. Borisova, R. Cao, H. Möhwald, D.Shchukin, Self-Healing and Antifouling Multifunctional Coatings Based on pHand Sulfide Ion Sensitive Nanocontainers, Advanced Functional Materials 23(26) (2013) 3307–3314.

[19] D.G. Shchukin, M. Zheludkevich, K. Yasakau, S. Lamaka, M.G.S. Ferreira, H.Möhwald, Layer-by-Layer Assembled Nanocontainers for Self-HealingCorrosion Protection, Advanced Materials 18 (13) (2006) 1672–1678.

[20] D.G. Shchukin, Container-based multifunctional self-healing polymer coatings,Polymer Chemistry 4 (18) (2013) 4871.

[21] A.S. Gnedenkov, S.L. Sinebryukhov, D.V. Mashtalyar, S.V. Gnedenkov, Localizedcorrosion of the Mg alloys with inhibitor-containing coatings: SVET and SIETstudies, Corrosion Science 102 (2016) 269–278.

[22] A.S. Gnedenkov, S.L. Sinebryukhov, D.V. Mashtalyar, S.V. Gnedenkov,Protective properties of inhibitor-containing composite coatings on a Mgalloy, Corrosion Science 102 (2016) 348–354.

[23] G.L. Li, M. Schenderlein, Y. Men, H. Möhwald, D.G. Shchukin, MonodispersePolymeric Core-Shell Nanocontainers for Organic Self-Healing AnticorrosionCoatings, Advanced Materials Interfaces 1 (1) (2014) 1300019.

[24] M. Sun, A. Yerokhin, M.Y. Bychkova, D.V. Shtansky, E.A. Levashov, A. Matthews,Self-healing plasma electrolytic oxidation coatings doped with benzotriazoleloaded halloysite nanotubes on AM50 magnesium alloy, Corrosion Science 111(2016) 753–769.

[25] D.N. Rockwood, R.C. Preda, T. Yücel, X. Wang, M.L. Lovett, D.L. Kaplan,Materials fabrication from Bombyx mori silk fibroin, Nature Protocols 6 (10)(2011) 1612–1631.

[26] C.B. Borkner, M.B. Elsner, T. Scheibel, Coatings and films made of silk proteins,ACS applied materials & interfaces 6 (18) (2014) 15611–15625.

[27] L. Zhou, X. Chen, Z. Shao, Y. Huang, D.P. Knight, Effect of Metallic Ions on SilkFormation in the Mulberry Silkworm, Bombyx m ori, The journal of physicalchemistry B 109 (35) (2005) 16937–16945.

[28] Differences in regenerated silk fibroin prepared with different solvent systemsFrom structures to conformational changes.pdf.

[29] H.J. Jin, J. Park, V. Karageorgiou, U.J. Kim, R. Valluzzi, P. Cebe, D.L. Kaplan,Water-Stable Silk Films with Reduced b-Sheet Content, Advanced FunctionalMaterials 15 (8) (2005) 1241–1247.

[30] L.-D. Koh, Y. Cheng, C.-P. Teng, Y.-W. Khin, X.-J. Loh, S.-Y. Tee, M. Low, E. Ye, H.-D. Yu, Y.-W. Zhang, M.-Y. Han, Structures, mechanical properties andapplications of silk fibroin materials, Progress in polymer science 46 (2015)86–110.

[31] B. Kundu, N.E. Kurland, S. Bano, C. Patra, F.B. Engel, V.K. Yadavalli, S.C. Kundu,Silk proteins for biomedical applications: Bioengineering perspectives,Progress in polymer science 39 (2) (2014) 251–267.

[32] M. Choi, D. Choi, J. Hong, Multilayered Controlled Drug Release Silk FibroinNanofilm by Manipulating Secondary Structure, Biomacromolecules 19 (7)(2018) 3096–3103.

[33] L. Meinel, D.L. Kaplan, Silk constructs for delivery of musculoskeletaltherapeutics, Advanced drug delivery reviews 64 (12) (2012) 1111–1122.

[34] S.L. McNamara, J. Rnjak-Kovacina, D.F. Schmidt, T.J. Lo, D.L. Kaplan, Silk as abiocohesive sacrificial binder in the fabrication of hydroxyapatite load bearingscaffolds, Biomaterials 35 (25) (2014) 6941–6953.

[35] F.M. Miroiu, G. Socol, A. Visan, N. Stefan, D. Craciun, V. Craciun, G. Dorcioman,I.N. Mihailescu, L.E. Sima, S.M. Petrescu, A. Andronie, I. Stamatin, S. Moga, C.Ducu, Composite biocompatible hydroxyapatite–silk fibroin coatings formedical implants obtained by Matrix Assisted Pulsed Laser Evaporation,Materials Science and Engineering: B 169 (1–3) (2010) 151–158.

[36] K. Numata, D.L. Kaplan, Silk-based delivery systems of bioactive molecules,Advanced drug delivery reviews 62 (15) (2010) 1497–1508.

[37] F. Mottaghitalab, H. Hosseinkhani, M.A. Shokrgozar, C. Mao, M. Yang, M.Farokhi, Silk as a potential candidate for bone tissue engineering, Journal ofcontrolled release : official journal of the Controlled Release Society 215(2015) 112–128.

[38] A.E. Thurber, F.G. Omenetto, D.L. Kaplan, In vivo bioresponses to silk proteins,Biomaterials 71 (2015) 145–157.

[39] Z. Li, X. Gu, S. Lou, Y. Zheng, The development of binary Mg-Ca alloys for use asbiodegradable materials within bone, Biomaterials 29 (10) (2008) 1329–1344.

[40] N. Li, Y.D. Li, Y.B. Wang, M. Li, Y. Cheng, Y.H. Wu, Y.F. Zheng, Corrosionresistance and cytotoxicity of a MgF2coating on biomedical Mg-1Ca alloy viavacuum evaporation deposition method, Surface and Interface Analysis 45 (8)(2013) 1217–1222.

[41] V. Di Castro, G. Polzonetti, XPS study of MnO oxidation, Journal of ElectronSpectroscopy and Related Phenomena 48 (1) (1989) 117–123.

[42] M.M. Chehimi, M. Delamar, X-ray photoelectron spectroscopy of merocyaninedyes: Part VIII. Partial charge and conjugation of heteroatoms in theelectroattractor rings11For preceding papers in this series, see refs. 1–7,Journal of electron spectroscopy and related phenomena 50 (2) (1990) C25–C32.

[43] O. Bayraktar, O. Malay, Y. Ozgarip, A. Batigun, Silk fibroin as a novel coatingmaterial for controlled release of theophylline, European journal ofpharmaceutics and biopharmaceutics : official journal ofArbeitsgemeinschaft fur Pharmazeutische Verfahrenstechnik e.V 60(3)(2005) 373-81.

[44] H.R. Bakhsheshi-Rad, M.H. Idris, M.R. Abdul-Kadir, Synthesis and in vitrodegradation evaluation of the nano-HA/MgF2 and DCPD/MgF2 compositecoating on biodegradable Mg–Ca–Zn alloy, Surface and Coatings Technology222 (2013) 79–89.

[45] Q. Lu, X. Hu, X. Wang, J.A. Kluge, S. Lu, P. Cebe, D.L. Kaplan, Water-insoluble silkfilms with silk I structure, Acta biomaterialia 6 (4) (2010) 1380–1387.

[46] J. Liu, B. Zheng, P. Wang, X. Wang, B. Zhang, Q. Shi, T. Xi, M. Chen, S. Guan,Enhanced in Vitro and in Vivo Performance of Mg-Zn-Y-Nd Alloy Achievedwith APTES Pretreatment for Drug-Eluting Vascular Stent Application, ACSapplied materials & interfaces 8 (28) (2016) 17842–17858.

[47] P. Wang, P. Xiong, J. Liu, S. Gao, T. Xi, Y. Cheng, A silk-based coating containingGREDVY peptide and heparin on Mg–Zn–Y–Nd alloy: improved corrosionresistance, hemocompatibility and endothelialization, Journal of MaterialsChemistry B 6 (6) (2018) 966–978.

[48] C. Jiang, X. Wang, R. Gunawidjaja, Y.H. Lin, M.K. Gupta, D.L. Kaplan, R.R. Naik, V.V. Tsukruk, Mechanical Properties of Robust Ultrathin Silk Fibroin Films,Advanced Functional Materials 17 (13) (2007) 2229–2237.

[49] C. Ding, Y. Liu, M. Wang, T. Wang, J. Fu, Self-healing, superhydrophobic coatingbased on mechanized silica nanoparticles for reliable protection of magnesiumalloys, Journal of Materials Chemistry A 4 (21) (2016) 8041–8052.

[50] N. Jadhav, C.A. Vetter, V.J. Gelling, The effect of polymer morphology on theperformance of a corrosion inhibiting polypyrrole/aluminum flake compositepigment, Electrochimica Acta 102 (2013) 28–43.

[51] L.-Y. Cui, S.-D. Gao, P.-P. Li, R.-C. Zeng, F. Zhang, S.-Q. Li, E.-H. Han, Corrosionresistance of a self-healing micro-arc oxidation/polymethyltrimethoxysilanecomposite coating on magnesium alloy AZ31, Corrosion Science 118 (2017)84–95.

[52] H. Zhang, R. Luo, W. Li, J. Wang, M.F. Maitz, J. Wang, G. Wan, Y. Chen, H. Sun, C.Jiang, R. Shen, N. Huang, Epigallocatechin gallate (EGCG) induced chemicalconversion coatings for corrosion protection of biomedical MgZnMn alloys,Corrosion Science 94 (2015) 305–315.

[53] Q.X. Ruan, P. Zhou, B.W. Hu, D. Ji, An investigation into the effect of potassiumions on the folding of silk fibroin studied by generalized two-dimensionalNMR-NMR correlation and Raman spectroscopy, The FEBS journal 275 (2)(2008) 219–232.

[54] H. Hornberger, S. Virtanen, A.R. Boccaccini, Biomedical coatings on magnesiumalloys - a review, Acta biomaterialia 8 (7) (2012) 2442–2455.

[55] S. Kunjukunju, A. Roy, M. Ramanathan, B. Lee, J.E. Candiello, P.N. Kumta, Alayer-by-layer approach to natural polymer-derived bioactive coatings onmagnesium alloys, Acta biomaterialia 9 (10) (2013) 8690–8703.

[56] A.S. Lammel, X. Hu, S.H. Park, D.L. Kaplan, T.R. Scheibel, Controlling silk fibroinparticle features for drug delivery, Biomaterials 31 (16) (2010) 4583–4591.

[57] Z. Jia, P. Xiong, Y. Shi, W. Zhou, Y. Cheng, Y. Zheng, T. Xi, S. Wei, Inhibitorencapsulated, self-healable and cytocompatible chitosan multilayer coating onbiodegradable Mg alloy: a pH-responsive design, Journal of MaterialsChemistry B 4 (14) (2016) 2498–2511.

[58] S. Midha, S. Murab, S. Ghosh, Osteogenic signaling on silk-based matrices,Biomaterials 97 (2016) 133–153.

[59] B. Liu, X. Zhang, G.Y. Xiao, Y.P. Lu, Phosphate chemical conversion coatings onmetallic substrates for biomedical application: a review, Materials science &engineering, C, Materials for biological applications 47 (2015) 97–104.



a ph-sensitive self-healing coating for biodegradable magnesium … · 2019-11-07 · self-healing...

Documents