Construction of highly conductive nanowires on a DNA template

Article abstract of DOI:10.1063/1.1338967

We present measurements of the electrical conductivity of metallic nanowires which have been fabricated by chemical deposition of a thin continuous palladium film onto single DNA molecules to install electrical functionality. The DNA molecules have been positioned between macroscopic Au electrodes and are metallized afterwards. Low-resistance electrical interfacing was obtained by pinning the nanowires at the electrodes with electron-beam-induced carbon lines. The investigated nanowires exhibit ohmic transport behavior at room temperature. Their specific conductivity is only one order of magnitude below that of bulk palladium, confirming that DNA is an ideal template for the production of electric wires, which can be utilized for the bottom-up construction of miniaturized electrical circuits.

Full text of DOI:10.1063/1.1338967

Construction of highly conductive nanowires on a DNA template
Jan Richter, Michael Mertig, Wolfgang Pompe, Ingolf Mönch, and Hans K. Schackert
Citation: Applied Physics Letters 78, 536 (2001); doi: 10.1063/1.1338967
View online: http://dx.doi.org/10.1063/1.1338967
View Table of Contents: http://scitation.aip.org/content/aip/journal/apl/78/4?ver=pdfcov
Published by the AIP Publishing
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
195.19.233.81 On: Sat, 04 Jan 2014 09:07:36

APPLIED PHYSICS LETTERS
VOLUME 78, NUMBER 4
22 JANUARY 2001
Construction of highly conductive nanowires on a DNA template
a)
Jan Richter, Michael Mertig, and Wolfgang Pompe
Institut f u¨ r Werkstoffwissenschaft, Technische Universit a¨ t Dresden, D-01069 Dresden, Germany
Ingolf M o¨ nch
Institut f u¨ r Festk o¨ rper- und Werkstofforschung, D-01171 Dresden, Germany
Hans K. Schackert
Abteilung Chirurgische Forschung, Universit a¨ tsklinikum Carl Gustav Carus der Technischen Universit a¨ t
Dresden, D-01069 Dresden, Germany
͑
Received 17 May 2000; accepted for publication 8 November 2000͒
We present measurements of the electrical conductivity of metallic nanowires which have been
fabricated by chemical deposition of a thin continuous palladium film onto single DNA molecules
to install electrical functionality. The DNA molecules have been positioned between macroscopic
Au electrodes and are metallized afterwards. Low-resistance electrical interfacing was obtained by
pinning the nanowires at the electrodes with electron-beam-induced carbon lines. The investigated
nanowires exhibit ohmic transport behavior at room temperature. Their specific conductivity is only
one order of magnitude below that of bulk palladium, confirming that DNA is an ideal template for
the production of electric wires, which can be utilized for the bottom-up construction of
miniaturized electrical circuits. © 2001 American Institute of Physics. ͓DOI: 10.1063/1.1338967͔
Building nanoscale structures with DNA is a focus of
current research, stimulated by the enormous progress in uti-
lizing DNA either for the controlled self-assembly of mo-
͑b͒ subsequent reduction of the complexes to form me-
tallic clusters.
In a typical experiment a Pd solution was prepared by
1
–7
lecularly designed nanostructures, or as template for thor-
dissolving 5 mg of Pd͑CH COO͒ in 1 ml HEPES buffer ͑pH
3
2
oughly engineered inorganic structures.6
,8–11
Nevertheless,
6
.5, 10 mM͒ by placing it in an ultrasonic bath for 2 min.
the construction of electronic circuits based only on native
DNA remains problematic, mainly due to the high resistance
of DNA that diminishes its potential applications in this
Afterwards it was centrifuged for 5 min at 2000 g to obtain a
saturated solution and to settle all undissolved particles. The
Pd solution was diluted 1:8 with buffer solution. A 2 ␮l drop
of DNA solution ͑0.1 mg/ml, ␭-DNA, New England Biolabs͒
was placed onto a gold contact structure. The positioning of
the DNA strands between the contacts is obtained by taking
advantage of a simple, but powerful physical alignment pro-
cess. The capillary forces applied by the receding front of the
evaporating drop containing the DNA molecules are used to
1
2–18
regard.
Recently, Braun and co-workers presented a new
approach by fixing DNA between two contacts and utilizing
6
it as template for the construction of a silver nanowire. This
technique uses the molecular recognition properties of the
molecule for the defined buildup of a nanostructure and in-
stalls its electrical functionality by the directed construction
of a metallic wire on the biotemplate. However, the reported
align the latter perpendicularly to the direction of the drying
100 nm thick silver wires displayed a electrical conductivity
23,24
front.
Some of them attach to the surface and are not
with a nonconducting gap for small bias voltages. Here, we
report the production of nanowires showing ohmic transport
behavior, which have been fabricated by direct growth of
palladium on a single DNA molecule. For nanowires with a
diameter above 50 nm, the specific conductivity is only one
order of magnitude below that of bulk palladium. These in-
vestigations demonstrate that metallized DNA molecules can
provide high conductive elements in a DNA-based circuitry.
In our experiment, we used metallized, linear ␭-DNA
molecules, 48 502 base pairs in length, to connect two gold
electrodes several micrometers apart, with the whole assem-
bly exhibiting an overall resistance well below 1 k⍀. DNA
going to be removed by the subsequent treatment. Then, a 25
l drop of the Pd solution was placed onto the sample with
the aligned DNA strands and reacted for 2 h at ambient con-
␮
ditions. Finally, a 10 ␮l drop of reduction solution containing
Ϫ1
Ϫ1
2
.5 gl sodium citrate, 2.5 gl of aqueous 85% lactic acid
Ϫ1
solution and 0.25 gl dimethylamine borane with pH 7.4
was added to initiate cluster growth. The samples were
rinsed carefully afterwards with aqua dest to avoid homoge-
neously grown clusters to settle down on the contact struc-
ture. The already grown cluster on the DNA can be devel-
oped by placing another drop of palladium solution on the
sample and adding reducing agent, etc.
metallization is accomplished in a two-step chemical depo-
sition of palladium,1
0,11,19,20
Initially, separated clusters are formed on the linear
involving
DNA template during the reduction process1
0,11
͓Figs. 1͑a͒
͑
a͒ activation of the template by treatment with Pd͑II͒
complexes, which in part bind on the DNA
and 1͑b͔͒ which is in close analogy to our observations for
the growth of palladium and platinum clusters on protein
2
1,22
strands,
and
templates such as bacterial surface layers and
microtubules.1
9,20
These clusters serve as a catalytic surface
a͒Electronic mail: jrichter@tmfs.mpgfk.tu.dresden
for the further reduction of palladium leading to a transition
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
003-6951/2001/78(4)/536/3/$18.00 536 © 2001 American Institute of Physics
95.19.233.81 On: Sat, 04 Jan 2014 09:07:36
0
1

Appl. Phys. Lett., Vol. 78, No. 4, 22 January 2001
Richter et al.
537
FIG. 1. Low voltage ͑1 kV͒ scanning electron microscope image of different
stages of the metallization process. ͑a͒ Linear chain of separated palladium
clusters connecting two gold contacts; ͑b͒ magnification of ͑a͒ showing
clusters with diameters up to 40 nm; ͑c͒ continuously coated DNA strand
after one development step with a diameter larger than 40 nm. The small
dendrites perpendicular to the strand arise from the growth of the metal film
on the template, that is driven by random processes.
FIG. 2. ͑a͒ Sketch of parallel oriented nanowires ͑vertical black lines͒ on the
comb-like gold contact structure ͑gray͒ showing five pairs of contact lines.
We used different types of samples from 5 to 10 ␮m spacing between the
gold contact lines. The alignment of DNA was achieved before metallization
by placing a drop of DNA solution on the contact structure and removing it
with filter paper perpendicular to the contact lines; ͑b͒ laser scanning micro-
scope image of a palladium wire. The nanowire bridges two gold lines that
are 10 ␮m apart. The left-hand side image is taken before and the right-hand
side one after the cutting of the nanowire using a standard micromanipulator
with etched tungsten tips.
from separated clusters into a continuous metal coating at a
minimum film thickness of about 20 nm after one or two
development steps ͓Fig. 1͑c͔͒.11 Transmission electron mi-
croscopy shows that the metal films exhibit a granular struc-
ture with grain sizes of 2–4 nm. By varying the number of
development steps, we are able to reproducibly build chains
of separated clusters through to wires with a continuous
metal coating. Thus, the approach presented here enables a
tunable metallization of DNA. In this letter, we will focus on
the electrical properties of continuously coated wires, since
they provide the reasonably high conductivity necessary for
electronic applications.
the palladium nanowire and the Au electrodes, could be con-
siderably improved by deposition of electron-beam-induced
carbon lines2
5–27
written over the very ends of the wires,
where they interface the Au electrodes. The impact of this
additional pinning is twofold. First, the wires are mechani-
cally fixed to the Au pads, and second, the deposition and
electron beam treatment lowers the contact resistance. The
pinning proved to be extremely effective. In most cases the
observed resistance drop after pinning was of the order of 5
k⍀, yielding overall two-terminal resistances below 1 k⍀,
and thus, maximum contact resistances of several hundred
⍀.
The described procedure allows the deposition of a large
number of nanowires with parallel orientation onto a comb-
shaped contact structure, where a number of wires bridge
two adjacent electrodes in a random manner ͓Fig. 2͑a͔͒, thus
providing a two-terminal measurement of an ensemble of
nanowires all connected in parallel. In this configuration, the
resistance R of one individual nanowire can be easily deter-
mined by breaking a particular wire by means of a microma-
nipulation device integrated in an optical microscope ͓see
Fig. 2͑b͔͒ and measuring simultaneously the overall resis-
tance of the system before and after cutting, Rb and Ra ,
respectively. In this case, the resistance of the wire removed
is given by RϭR R /(R ϪR ). The main advantage of the
Two-terminal I–V curves of single nanowires were
taken after removing all other wires from the sample. Figure
3 shows such a particular wire with a continuous Pd coating
around 50 nm in diameter. The corresponding I–V curve is
given in Fig. 4. We find ohmic behavior with a resistance of
only 743 ⍀. Cutting of this nanowire resulted in an insulat-
ing sample, proving that the measured conductance was in-
deed caused by this particular nanowire. The inset of Fig. 4
shows that the linear current–voltage dependence can be ob-
served for bias voltages down to 1 ␮V. No evidence of a
a
b
a
b
experimental approach described here is that it allows the
investigation of a large number of wires generated in a single
preparation step, and thus, enables the accumulation of sta-
tistically relevant datasets in a relatively short time.
6
28
nonconducting region or Coulomb blockade behavior, was
found at room temperature.
We investigated more than one hundred nanowires and
all showed ohmic behavior at room temperature, indicating
that a diameter of about 50 nm is sufficient to achieve a
continuous metallization of the DNA. However, although an
almost linear dependence of the resistance on the spacing
between adjacent Au electrodes was observed, indicating
that the wire resistance scales with its length, initially none
of the wires exhibited a resistance below 5 k⍀—even on the
assembly of relatively thick nanowires ͑diameters up to 200
The minimum specific conductivity, ␴ , of the nanowire
1
shown in Fig. 3 can be evaluated by neglecting the remaining
contact resistances between the wire and the Au electrodes.
4
Ϫ1
We calculate ␴ Ϸ2ϫ10 Scm for a wire length of 6.5 ␮m
1
and an average diameter of about 50 nm. This value is less
than one order of magnitude smaller than that of bulk palla-
dium giving clear evidence of the metallic character of the
wire conductivity. Furthermore, we estimated the specific
conductivity, ␴ , of a disordered Pd wire within the Drude
2
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
2
9
nm͒. This behavior, caused by contact resistances between
model assuming an electron mean free path of about 2 nm.
195.19.233.81 On: Sat, 04 Jan 2014 09:07:36

538
Appl. Phys. Lett., Vol. 78, No. 4, 22 January 2001
Richter et al.
nate the two-terminal resistance measured in our experiment,
illustrating the enormous benefit of the pinning procedure.
In conclusion, we have demonstrated the assembly of
highly conductive palladium nanowires on a DNA template.
The estimated specific conductance of the nanowires is only
one order of magnitude smaller than for bulk palladium.
Thus, the present study provides a step towards realistic elec-
tronic elements based on biological templates by showing
the template capabilities of DNA. Moreover, the metalliza-
tion of the DNA templates opens new possible applications
of DNA wiring at elevated temperatures where native, un-
coated DNA molecules are no longer stable.
This research is supported by the Deutsche Forschungsge-
meinschaft Grant No. Le 747/24.
1
N. C. Seeman, Annu. Rev. Biophys. Biomol. Struct. 27, 225 ͑1998͒.
A. P. Alivisatos, K. P. Johnsson, X. Peng, T. E. Wilson, C. J. Loweth, M.
2
P. Bruchez, and P. G. Schultz, Nature ͑London͒ 382, 609 ͑1996͒.
C. A. Mirkin, R. L. Letsinger, R. C. Mucic, and J. J. Storhoff, Nature
FIG. 3. Low voltage ͑1 kV͒ scanning electron microscope image of a single
palladium metallized DNA strand with a length of ϳ16 ␮m corresponding
to the length of a ␭-DNA molecule. The right-hand side of the strand con-
3
͑
London͒ 382, 607 ͑1996͒.
4
R. Elghanian, J. J. Storhoff, R. C. Mucic, R. L. Letsinger, and C. A.
Mirkin, Nature ͑London͒ 277, 1078 ͑1997͒.
R. F. Service, Science 277, 1036 ͑1997͒.
nects two gold electrodes over a SiO substrate. The inset shows a magni-
2
fication of the middle part with a diameter of 50 nm.
5
6
E. Braun, Y. Eichen, U. Sivan, and G. Ben-Yoseph, Nature ͑London͒ 391,
7
75 ͑1998͒.
7
8
C. A. Mirkin, MRS Bull. 25, 43 ͑2000͒.
J. L. Coffer, S. R. Bigham, X. Li, R. F. Pinizzotto, Y. Rho, G. Young, R.
M. Pirtle, and I. L. Pirtle, Appl. Phys. Lett. 69, 3851 ͑1996͒.
T. Torimoto, M. Yamashita, S. Kuwabata, T. Sakata, H. Mori, and H.
Yoneyama, J. Phys. Chem. B 103, 8799 ͑1999͒.
W. Pompe, M. Mertig, R. Kirsch, R. Wahl, L. C. Ciacchi, J. Richter, R.
Seidel, and H. Vinzelberg, Z. Metallkd. 90, 1085 ͑1999͒.
J. Richter, R. Seidel, R. Kirsch, M. Mertig, W. Pompe, J. Plaschke, and H.
K. Schackert, Adv. Mater. 12, 507 ͑2000͒.
M. R. Arkin, Science 273, 475 ͑1996͒.
Y. Okahata, T. Kobayashi, K. Tanaka, and M. Shimomura, J. Am. Chem.
Soc. 120, 6165 ͑1998͒.
J. Jortner, M. Bixon, T. Langenbacher, and M. E. Michel-Beyerle, Proc.
Natl. Acad. Sci. U.S.A. 95, 12759 ͑1998͒.
E. Meggers, M. E. Michel-Beyerle, and B. Giese, J. Am. Chem. Soc. 120,
The latter value corresponds to the typical grain size of the
Pd metallization film and provides an upper estimate of the
electron mean free path assuming electron scattering at grain
boundaries to be the dominant mechanism. These consider-
9
10
11
4
Ϫ1
ations yield ␴ Ϸ10 Scm in agreement with ␴ calculated
2
1
from experimental data. From these calculations we con-
clude that the remaining contact resistance does not domi-
12
13
1
1
1
4
5
6
12950 ͑1998͒.
P. J. de Pablo, P. Moreno-Herrero, J. Colchero, J. G. Herrero, P. Herrero,
A. M. Bar o´ , P. Ordej o´ n, J. M. Soler, and E. Artocho, Phys. Rev. Lett. 85,
4
992 ͑2000͒.
1
1
1
2
2
2
2
2
7
8
9
0
1
2
3
4
P. T. Henderson, D. Jones, G. Hampikian, Y. Kann, and B. G. Shuster,
Proc. Natl. Acad. Sci. U.S.A. 96, 8353 ͑1999͒.
D. Porath, A. Bezryadin, S. de Vries, and C. Dekker, Nature ͑London͒
4
03, 635 ͑2000͒.
M. Mertig, R. Kirsch, W. Pompe, and H. Engelhardt, Eur. Phys. J. D 9, 45
1999͒.
M. Mertig, R. Kirsch, and W. Pompe, Appl. Phys. A: Mater. Sci. Process.
96, 723 ͑1998͒.
͑
1
J. Duguid, V. A. Bloomfield, J. Benevides, and G. J. Thomas, Jr., Bio-
phys. J. 65, 1916 ͑1993͒.
G. B. Onoa, G. Cervantes, V. Moreno, and M. Prieto, Nucleic Acids Res.
26, 1473 ͑1998͒.
A. Bensimon, A. Simon, A. Chiffaudel, V. Croquette, and F. Heslot, Sci-
ence 265, 2096 ͑1994͒.
D. Bensimon, A. Simon, V. Croquette, and A. Bensimon, Phys. Rev. Lett.
7
4, 4754 ͑1995͒.
2
2
5
6
H. K o¨ nig and G. Helwig, Z. Phys. 129, 491 ͑1951͒.
A. N. Broers, W. W. Molzen, J. J. Cuomo, and N. D. Wittels, Appl. Phys.
Lett. 29, 596 ͑1976͒.
FIG. 4. Two-terminal current–voltage curves of the single, pinned nanowire
shown in Fig. 3 before ͑᭿͒ and after ͑᭡͒ cutting, taken with a patch-clamp
amplifier ͑EPC9, Heka, Germany͒. The wire resistance is 743 ⍀ correspond-
2
2
2
7
8
9
A. Bechtold, M. Henny, C. Terrier, C. Strunk, C. Sch o¨ nenberger, J.-P.
Salvetat, J.-M. Bonard, and L. Forro, Appl. Phys. Lett. 73, 274 ͑1998͒.
H. Grabert and M. H. Devoret, Single Charge Tunneling ͑Plenum, New
York, 1992͒.
4
Ϫ1
ing to a minimum estimated specific conductivity of ␴ Ϸ2ϫ10 S cm
.
1
The inset shows the ohmic I–V characteristic of the nanowire down to 1
V. After cutting, the sample was insulating.
C. Kittel, Introduction to Solid State Physics ͑Wiley, New York, 1953͒.
␮
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
95.19.233.81 On: Sat, 04 Jan 2014 09:07:36
1