NATURE PHOTONICS | VOL 4 | JUNE 2010 | www.nature.com/naturephotonics 349

news & views

require the quantum information contained in a fl ying photon qubit to be converted into a stationary qubit. Entanglement could then be created so that the quantum information is stored and at some point transmitted by reconversion into a photon. Th e spin of an electron may provide such a stationary memory qubit. Th e next challenge

in constructing a semiconductor-based quantum repeater will therefore be to establish entanglement between the qubits. ❐

Manfred Bayer is at the Institute of Physics, Dortmund University of Technology, Otto-Hahn-Strasse 4, Dortmund, D-44227, Germany. e-mail: manfred.bayer@physik.uni-dortmund.de

References

1. Berezovsky, J., Mikkelsen, M. H., Stoltz, N. G., Coldren, L. A. & Awschalom, D. D. Science 320, 349–352 (2008).

2. Press, D., Ladd, T. D., Zhang, B. & Yamamoto, Y. Nature 456, 218–221 (2008).

3. Greilich, A. et al. Nature Phys. 5, 262–266 (2009).4. Press, D. et al. Nature Photon. 4, 367–370 (2010).5. Greilich, A. et al. Science 313, 341–345 (2006).6. Hahn, E. L. Phys. Rev. 80, 580–594 (1950).

Optical tweezers rely on strongly focused laser beams to trap and move micro- and nanoscale

particles. To guarantee stable trapping, the gradient component of the light force must be greater than the scattering component, and thus high numerical aperture microscope objectives are used to bring the beam into sharp focus. Such optics are very well corrected, and form small diff raction-limited optical traps. However, heterogeneities in the refractive index of the surrounding sample are a source of signifi cant optical aberration and so greatly diminish the ability to form a sharp focal spot, thus making it diffi cult to achieve trapping in scattering media and thick biological tissue. Now, writing in Nature Photonics, Tomáš Čižmár and colleagues from the University of St Andrews in Scotland present an answer to this problem: an in situ method that compensates for strong aberrations, allowing optical trapping in a highly turbid medium for the fi rst time1.

Unwanted optical aberrations are a common problem in the deep imaging of biological specimens and in astronomy2,3. In the latter case, atmospheric turbulence causes images of stars captured by ground-based telescopes to be severely blurred. Th e common solution is to monitor the distortion and use an adaptive optical element, typically a deformable membrane mirror, to compensate for it. Th ese sample-induced optical aberrations not only deform the incoming wavefront, but also scatter light outside the region of interest. When attempting to achieve optical tweezing in a heterogeneous medium, the incident laser light suff ers from rapid spreading and dissipates to form a random

speckle pattern with low intensity. Stable trapping thus requires wavefront sensing prior to compensation to be performed very close to the target position. So far, most of the work on aberration correction in optical tweezers either has been concentrated on addressing the spherical aberration associated with oil-immersion

objectives, or is inappropriate for in situ correction in deep, complex samples4,5.

Th e research of Čižmár et al. follows from the work of Ivo Vellekoop and colleagues6–8, which demonstrated how to focus coherent light in strongly scattering media. Th e method uses a spatial light modulator as an adaptive optical element, and is based on the decomposition of the light fi eld into orthogonal modes. Th e active area of the spatial light modulator is divided into non-overlapping sub-domains with independent control; a probe senses the intensity at a target point, which is maximized by optimizing the amplitude and phase modulation at each sub-domain for the constructive interference of all modes. In the scheme of Čižmár et al., optimization is achieved by isolating two single modes at a time on the fi rst diff raction order, thus enhancing the contrast of interference signals. Th is enables the user to unlock the full benefi t of the detector’s dynamic range, and is especially advantageous when dealing with small aberrations. Čižmár et al. achieve improved results over Vellekoop and colleagues by performing complex compensation, rather than simply correcting the phase distortions. Care must be taken, however, in cases where many modes with small amplitudes may severely reduce the light effi ciency of the optical system.

Čižmár et al. demonstrate the potential of their correction method for optical trapping experiments1 (Fig. 1). Light from a fl uorescent or scattering metallic nanoparticle (both 200 nm in diameter) is used as a probe within the sample for optimizing aberration correction. Th ey stably trap 1 μm polystyrene beads at

WAVEFRONT CORRECTION

Trapping through turbid mediaThe aberrations induced by strongly scattering and turbid samples make optical trapping in such media impossible.

Now, researchers in Scotland have overcome the problem using in situ aberration correction.

Estela Martín-Badosa

Figure 1 | Optical aberrations in the laser

wavefront are compensated for by the spatial light

modulator, thus enabling stable trapping even after

propagation through a strongly scattering medium.

Spatial light

modulator

Microscope

objective

Scattering layerTrapped particle

run.indd 349run.indd 349 10.5.14 3:38:58 PM10.5.14 3:38:58 PM

© 20 Macmillan Publishers Limited. All rights reserved10

350 NATURE PHOTONICS | VOL 4 | JUNE 2010 | www.nature.com/naturephotonics

news & views

low power (0.2 mW) and with excellent trapping effi ciencies (trap stiff ness per unit laser power) of ~2.5 pN μm–1 mW–1 in the lateral direction and 0.5 pN μm–1 mW–1 in the axial direction, close to the theoretical values. Minimizing laser exposure is important in live biology experiments to reduce optical damage to the specimen.

Interestingly, the authors were also able to use the scheme to allow trapping using a ‘non-corrected’ high aperture condenser lens, instead of the usual well-corrected microscope objective, thus broadening the design possibilities of optical tweezers.

Th e most exciting and promising result, however, is successful trapping in a highly turbid medium. In this case, light propagating through a strongly scattering 40-μm-thick layer of PDMS (polydimethylsiloxane) fi lled with polystyrene beads (5 μm in diameter) is used to trap a 1.6 μm polystyrene bead suspended in water behind the scattering layer. A supplementary laser holds an identical bead, to be used as the intensity probe, at the exact trapping position. It is important to note that the aberration correction is site-specifi c, and must be repeated and modifi ed for every target point of the sample.

Th is constraint currently hinders the application of the method to real biological specimens, as individual particles functioning as sensing probes must be injected and immobilized at controlled sites within the sample, and this may not always be practical. More conveniently, the reference for optimization could be the target to be trapped itself, which should be static and labelled with a fl uorophore compatible with near-infrared, non-harmful optical manipulation. Quantum dots would probably be required, which, owing to their improved photostability and brightness over conventional organic dyes, can help in the detection of light from the probe and its diff erentiation from the background. Furthermore, owing to the long refresh time of the spatial light modulator technology, the optimization procedure is currently too slow to follow time-varying distortions (occurring in ex vivo or in vivo experiments). It is encouraging to hear that the authors are already considering an alternative design to achieve sub-second correction.

Th ere are other less demanding (yet still interesting) applications in which in situ aberration correction should yield signifi cant progress, such as in colloidal

studies, data storage and micromachining. For optical tweezers, trapping inside living cells has already been demonstrated9 in single-cell environments with small aberrations. Th e prospect of intact deep manipulation within complex tissues is both challenging and stimulating. Many other biological techniques, including nanosurgery, tissue engineering and advanced microscopy, would also benefi t from advances in in situ aberration correction. ❐

Estela Martín-Badosa is at the Department of Applied Physics and Optics, University of Barcelona, Martí i Franquès 1, E-08028 Barcelona, Spain. e-mail: estela.martinb@ub.edu

References

1. Čižmár, T., Mazilu, M. & Dholakia, K. Nature Photon. 4, 388–394 (2010).

2. Booth, M. J. Phil. Trans. R. Soc. A 365, 2829–2843 (2007).3. Girkin, J. M., Poland, S. & Wright, A. J. Curr. Opin. Biotechnol.

20, 106–110 (2009).4. Wulf, K. D. et al. Opt. Express 14, 4170–4175 (2006).5. Jesacher, A. et al. Opt. Express 15, 5801–5808 (2007).6. Vellekoop, I. M. & Mosk, A. P. Opt. Lett. 32, 2309–2311 (2007).7. Vellekoop, I. M., van Putten, E. G., Lagendijk, A. & Mosk, A. P.

Opt. Express 16, 67–80 (2008).8. Vellekoop, I. M. & Mosk, A. P. Phys. Rev. Lett. 101, 120601 (2008).9. Ashkin, A., Schütze, K., Dziedzic, J. M., Euteneuer, U. &

Schliwa, M. Nature 348, 346–348 (1990).

Although liquids typically fl ow downwards,

the capillary eff ect — a gravity-defying

process exploited by the roots and

stems of plants — also allows them to

be drawn up through narrow tubes. The

phenomenon relies on the adhesion of

liquid to the walls of a narrow vessel,

pulling the edges of the liquid up to form

a concave meniscus. Surface tension then

causes the entire column of liquid to be

drawn upwards — a process commonly

called wicking.

Now, A. Y. Vorobyev and Chunlei Guo

from the Institute of Optics at the

University of Rochester, USA, have

demonstrated that laser pulses can be

used to treat the surface of silicon so that

it becomes superhydrophilic and exhibits

the capillary eff ect (Opt. Express 18, 6455–

6460; 2010). The pulsed beam from an

amplifi ed Ti:Sapphire femtosecond laser

was used to create a 22 mm × 11 mm

array of parallel microgrooves on a

25 mm × 25 mm × 0.65 mm single-

crystal phosphorus-doped silicon sample.

Making silicon superhydrophilic

LASER FABRICATION

100 μm

The laser generated 65 fs pulses at a

central wavelength of 800 nm and with

1.5 mJ of energy per pulse, fabricating

each microgroove at a scanning speed

of 1 mm s–1 and with a focused laser spot

diameter of 100 μm. The resulting grooves

had an average depth of 40 μm.

The researchers tested the wetting

properties of the samples by applying

1–5 μL drops of various liquids, including

distilled water, acetone and methanol, to

their silicon surfaces. Using a camera to

capture the spreading dynamics of the

liquids at a speed of fi ve frames per

second, they were able to compare the

wicking behaviours of liquids on both

laser-treated and untreated samples

of silicon. The results showed that the

laser treatment caused the silicon to

become superhydrophilic, and that

the spreading distance of the liquids,

regardless of chemical composition,

followed the classical square root of time

dependence — a characteristic of the

Washburn equation, which governs the

motion of liquids in a closed capillary.

The same results were achieved when the

silicon samples were stood vertically, with

the liquid rising upwards against gravity.

Vorobyev and Guo believe that the

capillary properties of their ultrafast-

laser-treated silicon may fi nd applications

in a number of areas, including in

microfl uidics, lab-on-a-chip technology,

biomedicine, and chemical and

biological sensors.

SONIA SHAHI

UN

IVE

RS

ITY

OF

RO

CH

ES

TE

R

run.indd 350run.indd 350 10.5.14 3:38:58 PM10.5.14 3:38:58 PM

© 20 Macmillan Publishers Limited. All rights reserved10