Biomedical Materials and Devices
Magnesium as a Biodegradable and
Bioabsorbable Material for Medical
Implants
Harpreet S. Brar, Manu O. Platt, Malisa Sarntinoranont, Peter I. Martin, and Michele V. Manuel
For many years, stainless steel, co-
balt-chromium, and titanium alloys
have been the primary biomaterials
used for load-bearing applications.
However, as the need for structural ma-
terials in temporary implant applica-
tions has grown, materials that provide
short-term structural support and can
be reabsorbed into the body after heal-
ing are being sought. Since traditional
metallic biomaterials are typically bio-
compatible but not biodegradable, the
potential for magnesium-based alloys
in biomedical applications has gained
more interest. This paper summarizes
the history and current status of mag-
nesium as a bioabsorbable implant
material. Also discussed is the devel-
opment of a magnesium-zinc-calcium
alloy that demonstrates promising deg-
radation behavior.
INTRODUCTION
Biomaterial implants can either be
used to replace a diseased part or to as-
sist in the healing process. While the
former application requires implants to
stay in the body permanently, the latter
only requires that the implant remain
in the body temporarily. Thus, in situ-
ations where a permanent implant is
used for a short-term application, addi-
tional surgeries are required to remove
these devices once the healing process
is complete. This removal process in-
creases the cost of health care and
patient morbidity.1 In contrast, biode-
gradable materials dissolve after the
healing process is complete and thus,
no additional surgeries are required for
removal of these implants. This also
eliminates the complications associ-
ated with the long-term presence of
implants in the body. Lastly, once these
materials degrade within the body, it is
important that the degradation products
Vol. 61 No. 9 • JOM
How would you…
…describe the overall significance
of this paper?
Medical advances have significantly
increased the average life
expectancy and have resulted
in an ever increasing aging
population. Billions of dollars are
spent in cardiac and orthopedic
implants alone. However, most
of the implants being used today
are only biocompatible and
not bioabsorbable. Magnesium
combines bioabsorbability with
high specific strength, allowing for
the possibility to combine optimal
mechanical performance with
biodegradable and bioabsorbable
properties.
…describe this work to a
materials science and engineering
professional with no experience in
your technical specialty?
Recently, there has been an
increased interest in developing
biocompatible and bioabsorbable
materials for numerous medical
applications ranging from cardiac
stents to orthopedic devices. This
paper provides an overview of the
history of the use of magnesium
as a biomaterial, biological
considerations in designing
bioabsorbable magnesium-based
materials, and compares the in vitro
performance of a new magnesium-
zinc-calcium alloy against pure
magnesium and commercially
available AZ91.
…describe this work to a
layperson?
Biomedical implants are required
to aid in the repair or replacement
of damaged body parts. Although
there is a specific need for temporary
implants, the majority of traditional
metallic biomaterials are designed
for long-term use. Bioabsorbable
implants provide short-term
support and can be easily absorbed
biologically. Because of its ability
to be metabolized by the body,
magnesium shows great promise
as a bioabsorbable implant material.
www.tms.org/jom.html
Overview
are able to be metabolized by the body,
and thus are bioabsorbable.2,3
Polymers were the first materials to
be used as commercial biodegradable
and bioabsorbable implant materials.
The earliest and most commonly used
absorbable materials include poly-
glycolic acid (PGA), poly-lactic acid
(PLA), and poly-dioxanone (PDS).4,5
However, these materials are limited
by their low mechanical properties and
radiolucency.6 Low strength severely
restricts the applications of polymeric
materials in load-bearing and tissue-
supporting applications, as a greater
amount of material is required to meet
the mechanical needs of the body. It
has also been observed that radiolucent
polymer stents can decrease the accu-
racy during the positioning of coronary
stents.6 Metals have desirable mechani-
cal properties due to their relatively
high strength and fracture toughness;
however, the majority of metals are
biologically non-absorbable or toxic.
Studies have shown that conventional
surgical alloys, like stainless steel, co-
balt, chromium, and nickel-based al-
loys produce corrosion products, which
are harmful to the human body.7–10 On
the other hand, magnesium (Mg) and
its corrosion products have excellent
biocompatibility and are considered to
be a promising technology for tempo-
rary medical implants.8,11 As a result,
the use of Mg as a biodegradable and
bioabsorbable medical material has
gained significant attention in the area
of biomaterials.
MAGNESIUM AS A
BIOMATERIAL
Magnesium shows great promise as a
potential biocompatible and biodegrad-
able material. Some of the attractive
physical characteristics of Mg include
31

Table I. Summary of Mechanical Properties of Natural and Implant Materials
Material
Natural Materials
Collagen
Cortical bone
Inorganic Materials
Magnesium
Stainless steels
Cobalt alloys
Titanium alloys
Platinum alloys
Synthetic hydroxyapatite
Organic Materials
L-PLA
D,L-PLA
UHMWPE
*Indicates compressive strength (MPa)
(a) Reference 36; (b) Reference 37; (c) Reference 30; (d) Reference 15
high specific strength and an elastic
modulus that is closest to the human
bone when compared to traditional me-
tallic implant materials (Table I). These
properties are of great importance as
high mechanical strength reduces the
amount of implant material needed for
a given applied load and reducing the
elastic modulus mismatch alleviates
stress-shielding effects between bone
and the implant material.
Magnesium is an essential mineral
for human metabolism and its deficien-
cy has been linked to various patholog-
ical conditions.7 The human body con-
tains about 1 mole (24 g) of Mg. It is
the second most common intracellular
ion and serves as a cofactor for more
than 300 enzymatic reactions ranging
from muscle contraction to neuronal
control.12 Most of the Mg in the body
(53%) is stored in bone, in an apatite
inorganic matrix subject to regulated
release.13 Magnesium-based materials
were first introduced for orthopedic ap-
plications in the beginning of the 20th
century. Lambotte14,15 first reported the
use of a pure Mg plate along with gold-
plated steel nails to secure a lower leg
bone fracture. However, the in vivo
corrosion of the implant was too rap-
id as it degraded in just 8 days. Since
then, several attempts have been made
to increase the corrosion resistance of
Mg implants and a variety of new al-
loys have been tested in vitro and in
vivo.16–18 These studies highlighted
Mg’s additional beneficial attributes,
namely its ability to stimulate bone
32
substrate on the implant comprised of
the proteins necessary for their function
Tensile Elastic and survival. For example, osteoblasts
Strength Modulus
(MPa) (GPa) Bio-degradable are bone-forming cells that lay down
the tough protein collagen and then
mineralize it to make new bone. Grow-
60 b 1b Yes
100–200 b 10–20 b Yes ing these cells on collagen increases
their bone mineralization activity.27
185–232c 41–45d Yes Alternatively, an elastic protein, aptly
480–834 b 193 b No named “elastin,” is abundant in arter-
655–1400 b 195–210 b No ies to expand and recoil when the large
550–985 b 100–105 b No volume of blood comes from heart con-
152–485 b 147 b No
600 d,* 73–117d Variable traction.28 After the cells are supported,
absorption of the implant would leave
28–50 a 1.2–3 a Yes behind a naturally synthesized protein
29–35 a 1.9–2.4 a Yes structure appropriate for those cells at
39–40 b 0.94–1.05 b No that specific site. For these reasons, a
significant, uncontrolled, local change
in Mg concentration due to implant
degradation can have a deleterious ef-
growth and healing. Later experiments fect on human physiology and must be
have confirmed that the presence of Mg managed through proper engineering
enhances the bone cell adhesion on alu- design. Hence for extensive use of Mg
mina19 and has no inhibitory effect on and its alloys in biomedical implant ap-
cell growth.20 Furthermore, corrosion plications, understanding and control
and degradation of Mg leads to the for- over the degradation rate are required.
mation of harmless corrosion products,
CONTROLLED
which are excreted through urine.21
DEGRADATION AND
However, the major limitation of Mg
ALLOY DESIGN
is its low corrosion resistance. High
corrosion rates result in the rapid re- Although the general biocompat-
lease of degradation products. A high ibility of Mg is high, increased deg-
rate of degradation under physiological radation rates under physiological pH
conditions can cause a reduction in the conditions can locally reduce the bio-
mechanical integrity of the implant be- compatibility near the implant surface.
fore the bone or tissue has sufficiently Under typical atmospheric conditions,
healed.15 Magnesium’s low corrosion Mg reacts with water to produce a
resistance also leads to the rapid pro- mildly protective film of magnesium
duction of hydrogen gas and the for- hydroxide (Mg(OH)2).29 Although this
mation of gas bubbles. These bubbles film slows corrosion under aqueous
can accumulate around the implant and conditions, it reacts with chlorine ions
delay the healing of the tissue.15 The lo- present in physiological conditions to
calized formation of hydrogen gas can produce MgCl2 and hydrogen gas.15
also result in a pH increase around the Efforts to control the corrosion rate
implant.22 This can cause local alkali- of Mg have utilized various processing
zation and severely affect the pH-de- methods such as purification, alloying,
pendent physiological processes in the anodizing, and surface coating. Studies
vicinity of the implant.23 have shown that purification of Mg re-
To successfully employ bioabsorb- duces the corrosion rate considerably;
able metal implants, the time frame however, due to the low yield strength
of degradation must be sufficient such of pure Mg,30 its application in orthope-
that the cells can synthesize and depos- dics and other load bearing applications
it an extracellular matrix for their own is limited.23 Alloying elements can be
support and function before the struc- added to increase the strength of pure
tural integrity of the implant is com- Mg31 but alloying elements should be
promised. Surfaces have been treated selected carefully to maintain the Mg’s
or coated with a variety of chemistries biocompatibility. Elements like Fe, Ni,
and polymers to encourage cell attach- Cu, and Co have extremely deleterious
ment.24–26 Once adhered, cells create a effects on the corrosion properties of
www.tms.org/jom.html JOM • September 2009
Mg, rapidly increasing the degrada-
tion rate. Cadmium, Mn, Sn, Zn, and
Ca have a mild effect on the corrosion
rate of Mg with their efficacy being
dependent on solute concentration.29
Aluminum is a major alloying element
in Mg-based alloys and is considered to
enhance the strength and corrosion re-
sistance. However, its poor biocompati-
bility causes phosphate depletion in tis-
sues and lowers phosphate absorption
from the digestive tract. The depletion
of phosphates can lead to progressive
senile dementia.32 Zirconium, which is
added as a grain refiner in Mg-based al-
loys, has been linked to breast and lung
cancer, and rare earth elements such as
Ce, Lu, and Pr are generally considered
toxic for the human body.23 Hence,
the selection of appropriate alloying
elements is critical in the design of a
biocompatible implant material. Using
materials essential to the human body
as alloying elements can greatly reduce
the chance of toxicity, allowing for the
development of a completely biode-
gradable and biocompatible alloy.
Calcium and Zn are two essen-
tial elements in human body that also
provide mechanical strengthening in

DEGRADATION STUDY OF PURE Mg AND Mg-Zn-Ca ALLOY

In vitro testing was performed on Mg-
15wt.%Zn-2wt.%Ca (ZX152) and com- Table A. Ion Concentrations of Simulated Body Fluid (SBF)34 and Hank’s Solution
mercially pure Mg (99.95%) samples. Disc- Ion Concentration
shaped samples were cut from cylindrical (mmol/L) Na+ K+ Ca2+ Mg2+ HCO3– Cl– HPO42– SO42–
rods. The samples were then mounted in ep-
oxy and polished up to 0.3 Mm. The mounted SBF 142.0 5.0 2.5 1.5 4.2 148.5 1.0 0.5
samples were immersed in Hank’s solution at Hank’s Solution 141.4 5.8 — — 4.2 142.2 0.8 —
37C for different times ranging from 30 min.
to 4 days. Samples were removed from the solution and washed cm2 in the ZX152 alloy.
with ethanol. Inductively coupled plasma (ICP) spectroscopy was The SEM micrographs of the cross section of the Mg and ZX152
then used to calculate the amount of Mg, Ca, and Zn present in the samples immersed after 219 h are shown in Figure B. It can be
Hank’s solution. Some of the samples were cross-sectioned to ob- seen that the morphologies of the interfaces between the corrosion
serve the interface between the corrosion layer and metal substrate layer and the substrate are different in pure Mg and ZX152. Pure
using scanning electron microscopy (SEM). Mg displays uniform surface degradation whereas site-specific
The degradation rates of pure Mg, Mg-Zn-Ca, and Mg- pitting can be seen in the ZX152 alloy. This is mainly due to the
9wt.%Al–1wt.%Zn (AZ91) are shown in Figure A. The dissolu- microstructural differences as the ZX152 alloy contains a eutectic
tion rates for AZ91 were obtained from Xin et al.34 who performed phase whereas pure Mg is single phase. In the ZX152 alloy, the
corrosion testing of the alloy in simulated body fluid (SBF). The eutectic phase seems to be more corrosion resistant as compared
ion concentrations for Hank’s solution and SBF are shown in to the matrix, thereby resulting in the preferential corrosion of the
Table A. The relative ion concentrations for both solutions are Mg-rich matrix over the eutectic. Xin et al.34 made a similar obser-
similar except that Hank’s solution does not contain Mg, Ca, or vation in AZ91 where B-Mg17Al12 shows a more passive behavior
sulfate ions prior to sample immersion. Although this allows for than the matrix. It has been suggested that B-Mg17Al12 affects cor-
increased resolution in the ICP measurements, the large concen- rosion in two ways.35 Apart from acting as a corrosion barrier, it
tration gradients between the sample surface and the solution may acts as a cathode for the Mg-rich matrix, thereby creating a gal-
increase the degradation rate relative to the same sample tested vanic couple with the matrix and increasing the corrosion rate.
in SBF. Additionally, the increased chlorine ion concentration in Making a similar conclusion about the cathodic nature of eutectic
the Hank’s solution is expected to further increase the degradation phase in ZX152 is beyond the scope of this paper. However, plots
rate relative to sample immersion in SBF. depicting degradation rates suggest that ZX152 alloy has a lower
Both pure Mg and AZ91 show a rapid degradation rate during degradation rate than AZ91 and pure Mg.
the first 10 h of immersion. Initial degradation rates for AZ91 and
pure Mg are approximately 0.15 and 0.11 mg/h-cm2, respectively.
This is contrasted with the lower degradation rate of 0.06 mg/h-
Figure B. An SEM mi-
0.16 crograph highlighting
the morphology of the
0.14 corrosion layer forming
Degradation Rate (mg/h-cm2)

on a ZX152 sample af-

0.12 ZX152 ter 219 hours of expo-
Mg sure to Hank’s solution.
0.10
AZ91 The eutectic (light col-
0.08 ored phase) phase can
be seen acting as a cor-
0.06 rosion barrier.

0.04

0.02
Figure C. An SEM mi-
0 crograph highlighting
0 5 10
25 30 35 40 45 5015 20 the morphology of the
Time (h) corrosion layer forming
Figure A. A plot of degradation rate versus time for three materi- on a pure Mg sample
als: Mg-15wt.%Zn-2wt.%Ca (ZX152), pure Mg, and Mg-9wt.%Al- after 219 hours of expo-
1wt.%Zn (AZ91). sure to Hank’s solution.

Vol. 61 No. 9 • JOM www.tms.org/jom.html 33

 27–37. 26. M.O. Platt et al., Journal of Cellular Physiology (18
Mg-based alloys. Calcium has been re- 8. R.C. Zeng et al., Advanced Engineering Materials, June 2009), doi: 10.1002/jcp.21854.
ported to improve the corrosion resis- 10 (8) (2008), pp. B3–B14. 27. R.M. Salasznyk et al., Journal of Biomedicine &
tance of Mg-based alloys in simulated 9. D. Upadhyay et al., Materials Science and Engineer- Biotechnology, 2004 (1) (2004), pp. 24–34.
ing A—Structural Materials Properties Microstructure 28. D.Y. Li et al., Nature, 393 (6682) (1998), pp. 276–
body fluid.33 Meanwhile, Zn additions and Processing, 432 (1-2) (2006), pp. 1–11. 280.
increase the strength of Mg-based al- 10. P.A. Dearnley, Surface & Coatings Technology, 198 29. B.A. Shaw, ASM Handbook—Corrosion: Funda-
loys primarily through precipitation (1-3) (2005), pp. 483–490. mentals, Testing, and Protection (Materials Park, OH:
11. B. Heublein et al., Heart, 89 (2003), pp. 651–656. ASM International, 2003), pp. 692–696.
strengthening. Thus a new Mg-Zn-Ca 12. D. MacLeod and J. MacLeod, British Journal of An- 30. C.J. Smithells, E.A. Brandes, and G.B. Brook,
alloy has been designed for potential aesthesia, 83 (6) (1999), pp. 972–973. Smithells’ Metals Reference Book. 7th ed. (London:
use as a biodegradable and bioabsorb- 13. R.J. Elin, Clinical Chemistry, 33 (11) (1987), pp. Butterworths, 1992), 1 v. (various pages).
1965–1970. 31. C.S. Roberts, Magnesium and its Alloys, Wiley Se-
able implant material. See the sidebar 14. A. Lambotte, Bull. Mem. Soc. Nat. Chir., 28 (1932), ries on the Science and Technology of Materials (New
for a presentation of our study on the pp. 1325-1334. York: Wiley, 1960).
degradation behavior of this alloy. 15. M.P. Staiger et al., Biomaterials, 27 (9) (2006), pp. 32. T.D. Luckey and B. Venugopal, Metal Toxicity in
1728–1734. Mammals (New York: Plenum Press, 1977).
CONCLUSION 16. V.V. Troitskii and D.N. Tsitrin, Khirurgiia, 8 (1944), 33. M.B. Kannan and R.K.S. Raman, Biomaterials, 29
pp. 41–44. (15) (2008), pp. 2306–2314.
Although a substantial amount of 17. E.D. McBride, Journal of the American Medical As- 34. Y.C. Xin et al., Journal of Materials Research, 22
sociation, 111 (1938), pp. 2464–2466. (7) (2007), pp. 2004–2011.
research has been performed on the 18. F. Witte et al., Biomaterials, 26 (17) (2005), pp. 35. G.L. Song and A. Atrens, Advanced Engineering
use of Mg and its commercially avail- 3557–3563. Materials, 1 (1) (1999), pp. 11–33.
able alloys for biomedical applications, 19. H. Zreiqat et al., Journal of Biomedical Materials 36. I. Engelberg and J. Kohn, Biomaterials, 12 (3)
Research, 62 (2) (2002), pp. 175–184. (1991), pp. 292–304.
further research is needed to fully 20. L. Li, J. Gao, and Y. Wang, Surface and Coatings 37. J. Black, Orthopaedic Biomaterials in Research
evaluate the potential of research-grade Technology, 185 (1) (2004), pp. 92–98. and Practice (New York: Churchill Livingstone, 1988).
Mg-based alloys for use in biologi- 21. N.E.L. Saris et al., Clinica Chimica Acta, 294 (1-2)
(2000), pp. 1–26. Harpreet S. Brar, Peter I. Martin, and Michele V. Man-
cal implants. An integrated approach 22. G.-L. Song and S.-Z. Song, Acta Physico-Chimica, uel are with the Materials Science and Engineering
involving chemists, experimental and 22 (10) (2006), pp. 1222–1226. Department, University of Florida, Gainesville, FL;
computational material scientists, and 23. G. Song, Corrosion Science, 49 (4) (2007), pp. Manu O. Platt is with the Wallace H. Coulter Depart-
1696–1701. ment of Biomedical Engineering, Georgia Tech and
medical professionals is needed to un- 24. C.D. Reyes, T.A. Petrie, and A.J. Garcia, Journal of Emory University, Atlanta, GA; and Malisa Sarn-
derstand the fundamental mechanisms Cellular Physiology, 217 (2) (2008), pp. 450–458. tinoranont is with the Department of Mechanical
involved with the application of Mg 25. T.A. Petrie et al., Journal of Cellular and Mo- and Aerospace Engineering, University of Florida,
lecular Medicine (August 2008), doi 10.1111/j.1582- Gainesville, FL. Dr. Manuel can be reached at (352)
alloys in physiological conditions. Un- 4934.2008.00476.x 846-3780; e-mail mmanuel@mse.ulf.edu.
derstanding the fundamental questions
like the effect of secondary phases on
the degradation rate, interaction of cor-
rosion surface with tissue, influence of
degradation products on surroundings,
and toxicity of alloying elements is
necessary to achieve the goal of a fully
biocompatible and biodegradable Mg
alloy implant.
ACKNOWLEDGEMENTS
The authors would like to gratefully
acknowledge Anil Sachdev and Jason
Traub from General Motors Corpora-
tion for their assistance in the process-
ing and preparation of the materials
used in this study.
References
1. J.B. Park and Y.K. Kim, Biomaterials: Principles and
Applications, ed. J.B. Park and J.D. Bronzino (Boca
Raton: CRC Press, 2003), pp. 1–20.
2. H. Hamid and J. Coltart, McGill Journal of Medicine,
10 (2) (2007), pp. 105–111.
3. P. Peeters et al., Journal of Endovascular Specialists,
12 (2005), pp. 1–5.
4. J. Heller, R.V. Sparer, and G. Zentner, Biodegradable
Polymers as Drug Delivery Systems, ed. M. Chasin
and R.S. Langer (New York: M. Dekker, 1990), pp.
121–161.
5. O.M. Bostman, Journal of Bone and Joint Surgery,
73B (4) (1991), p. 682.
6. J. Levesque et al., Advanced Materials & Processes,
162 (9) (2004), pp. 45–48.
7. J. Vormann, Mol Aspects Med, 24 (1-3) (2003), pp.

34 www.tms.org/jom.html JOM • September 2009