Supporting Information -

Temperature-Responsive Nanomagnetic Logic

Gates for Cellular Hyperthermia

Rui Oliveira-Silva,†,‡ Rute A. Pereira,† Fabio M. Silva,† Vıtor M. Gaspar,¶

Alfonso Ibarra,§ Angel Millan,‖ Filipa L. Sousa,¶ Joao F. Mano,¶ and Nuno J.

O. Silva∗,†

†Departamento de Fısica and CICECO, Aveiro Institute of Nanotechnology Universidade

de Aveiro, 3810-193 Aveiro, Portugal.

‡Univ. Lisbon, Inst. Super. Tecn., Inst. Bioengn. & Biosci., Dept. Bioengn., Ave Rovisco

Pais, P-1049001 Lisbon, Portugal.

¶Departamento de Quımica and CICECO, Aveiro Institute of Nanotechnology,

Universidade de Aveiro, 3810-193 Aveiro, Portugal.

§LMA, Instituto de Nanociencia de Aragon, Universidad de Zaragoza.

‖Instituto de Ciencia de Materiales de Aragon, CSIC - Universidad de Zaragoza.

Departamento de Fısica de la Materia Condensada, Facultad de Ciencias, 50009 Zaragoza,

Spain.

E-mail: nunojoao@ua.pt

Detailed experimental description

Here we detail the synthesis and experiments performed in four samples of Fe3Se4 nanoplatelets,

whose general procedure is given in the manuscript. Specific conditions and selected prop-

1

Electronic Supplementary Material (ESI) for Materials Horizons.This journal is © The Royal Society of Chemistry 2018

erties are given in table 1.

Synthesis of precursors

Fe-oleate was prepared and washed using standard procedures .1 Washing processes included

an extraction with a mixture of ethanol acetone and a redispersion with hexane followed by

an evaporation step at room temperature. These steps were repeated form 2 to 4 times,

resulting in brown Fe-oleates spanning from a slightly viscous behaviour to a solid darker

wax, used to produce smaller and larger nanoplatelets, respectively. The Se-octadecene

complex was also prepared using standard procedures2 by the reaction of 434 mg of metallic

Se powder with 75.3 g of 1-octadecene during 30 min at 180 ◦C.

Microscopy

Transmission electron microscopy (TEM) on the nanoparticles was performed using a Hitachi

H9000 microscope, working at 300 kV. Scanning Transmission Electron Microscopy - High

Angle Annular Dark Field (STEM-HAADF) images were obtained in a probe-corrected Titan

(FEI) at a working voltage of 300 kV, coupled with a HAADF detector (Fischione). In order

to analyse the chemical composition of the materials, X-ray Energy Dispersive Spectra (EDS)

were obtained with an EDAX detector. Samples for TEM observations of the NPs dispersed

in the ferrofluids were prepared by dip coating of carbon coated copper grids.

Prostate cancer cells were washed 3 times with PBS and imaged in a Zeiss Imager M2

fluorescence microscope (Carl Zeiss Microscopy GmbH, Germany), equipped with a 10x/0.25

Plan Apochromat air objective. All the data was processed in the Zeiss Zen software (v 2.3).

X-ray diffraction

X-ray diffraction (XRD) measurements were performed at room temperature with a PANa-

lytical Empyrean powder diffractometer using monochromated CuKα radiation (λ = 1.541

2

A) in the 10 - 80◦ 2θ range at 0.02◦ resolution, and 4000 acquisition points per step. The

incident beam optics included a Soller slit of 0.04 rad, a 10 mm fixed mask, a divergence

fixed slit of 1/4 and an anti-scatter slit of 1/8. The diffracted beam optics included a Soller

slit of 0.04 rad and anti-scatter slit of 7.5 mm. The analysis of the diffraction patterns was

performed by Rietveld refinement using the FullProf package.3 The size effects were treated

with the integral breadth method using the Voigt model for both the instrumental and intrin-

sic diffraction peak shape considering a Thompson-Cox-Hastings pseudo-Voigt convoluted

with Axial divergence asymmetry function to describe the peak shape. The contribution

of the instrument to the peaks broadening was determined by the refinement of the XRD

pattern of a LaB6 standard sample (NIST ref. 660a).

Magnetic data

Magnetic measurements were performed in a superconducting quantum interference device

(SQUID) magnetometeres model MPMS-XL and MPMS3, from Quantum Design Inc, under

helium atmosphere and under controlled magnetic field and temperature. The field is pro-

vided by a superconducting magnet and temperature is controlled in feedback using a cold

helium flow and a heater acting on the flow

X-ray photoelectron spectroscopy

The X-ray photoelectron spectroscopy (XPS) analysis was carried out using a Kratos Axis

SUPRA spectrometer employing a monochromatic Al Kα (1486.6 eV) 15 mA, 15 kV) X-ray

source and a power of 225 W.

Cell culture

Prostate cancer cells (PC-3) were routinely cultured in RPMI-1640 medium supplemented

with 10% FBS and 1% antibiotic/antimycotic. PC-3 cells were grown in cell culture treated t-

3

flasks in a temperature-controlled incubator at 37◦, 5% CO2 and in a humidified atmosphere.

Upon reaching confluency cells were detached by using TripLETMXpress and subcultured.

Hyperthermia studies on cells

The normal use of the nanoplatelets as logic gates, to know if a temperature threshold

during irradiation is crossed, involves setting them in a high magnetization state (M=1) by

the application of a (high) magnetic field (H=1), followed by the irradiation procedure and

followed by measuring the final magnetization state, which may or may not have changed to

zero due to irradiation). Although less useful in practice, the use of the SR latch gate can

be reversed and used to detect any magnetic field going above a threshold during a period

of time where temperature is known to be always below TC . In this case, temperature is

turned first such that magnetization is reset low (M=0) and any possible field during a given

period of time will set magnetization in a high state (Q=1).

Simulations

Thermal simulations were performed using the finite element analysis software QuickFieldTMProfessional

6.3. We considered a axial symmetric geometry with hot spots representing the nanoplatelets

at a constant temperature (42 ◦C) embedded in a 780µm thick MatrigelTMlayer with ther-

mal conductivity 0.5 W/Km and a 20µm thick cell layer with the same conductivity. The

top cell layer exchanges heat with the cell medium by convection, simulated using a surface

convection boundary condition with h = 500 W/Km2 and ambient temperature 37 ◦C.4 The

remaining MatrigelTMand cell surfaces exchange heat by conduction with a plastic frame

(thermal conductivity 0.4 W/Km), which in turn exchanges heat with air by convection

(h = 10 W/Km2 and ambient temperature 37 ◦C). This simulation provides an estimation

of the maximum expected temperature difference between the nanoplatelets and the cells

due to the chosen geometry. Additional temperature drops may occur at the micro/nano

interfaces and are presently a matter of debate (see, for instance, Ref.5). Such additional

4

temperature drops would be present even if the nanoplatelets were mixed together will cells.

References

(1) L. M. Bronstein, X. Huang, J. Retrum, A. Schmucker, Maren Pink, Barry D. Stein, and

Bogdan Dragnea. Influence of iron oleate complex structure on iron oxide nanoparticle

formation. Chem. Mater., 19:3624–3632, 2007.

(2) Craig Bullen, Joel van Embden, Jacek Jasieniak, Joanna E. Cosgriff, Roger J. Mul-

der, Ezio Rizzardo, Min Gu, and Colin L. Raston. High activity phosphine-free sele-

nium precursor solution for semiconductor nanocrystal growth. Chemistry of Materials,

22(14):4135–4143, 2010.

(3) J. Rodrıguez-Carvajal. Physica B, 192:55, 1993.

(4) Sean D. Lubner, Jeunghwan Choi, Geoff Wehmeyer, Bastian Waag, Vivek Mishra, Har-

ishankar Natesan, John C. Bischof, and Chris Dames. Reusable bi-directional 3 sensor

to measure thermal conductivity of 100-m thick biological tissues Review of Scientific

Instruments, 86:014905, 2015.

(5) G. Baffou, H. Rigneault, D. Marguet, L. Jullien. A critique of methods for temperature

imaging in single cells. Nat. Methods, 11(9):899-901, 2014.

Tables and figures

5

Table 1: Relevant parameters of synthesis conditions, morphology and magnetic propertiesof the four samples of Fe3Se4 nanoplatelets presented in the manuscript. The diameter of thenanoplatelets is estimated as the distance across two parallel edges. Dg is calculated usingthe average diameter and thickness considering that the nanoplatelets are flat cylinders.

synthesis conditions characteristic sizes and magnetic phase transitionramp rate Tmax time @Tmax diameter thickness XRD size Dg TC

(◦C/min) (◦C) (h) (nm) (nm) (nm) (nm) (◦C)7.5 210 1 130 43 18 95 185 210 1 95 36 26 70 3618 210 0.75 193 46 21 137 2218 210 1 250 60 31 180 42

Table 2: True table of the SR latch as usually presented, showing the output of the logiccircuit at a time t (Q(t)), the subsequent action on S, R, or both and the resulting output ata time t+ 1 (Q(t+ 1)). RC stands for ’race condition’, where the state Q(t+ 1) is ill-definedand depend on which S or R is set low in first place. In the present case, Q corresponds tothe magnetization M, S corresponds to the field H and R corresponds to temperature T.

Q(t) S R Q(t+ 1)0 0 0 00 0 1 00 1 0 10 1 1 RC1 1 1 RC1 0 0 11 0 1 01 1 0 1

6

Figure 1: Electron microscope images of Fe3Se4 nanoplatelets. Left column showsimages of sample with average lateral diameters of 130 nm, central column 95 nm and rightcolumn 193 nm. All scale bars correspond to 500 nm.

80 100 120 140 160 1800

20

40

60

60 80 100 120 140 100 150 200 250 180 200 220 240 260 280 300 320 340

20 30 40 50 60

0

10

20

30

40

50

60

70

80

20 30 40 50 60 70 20 40 60 80 100 20 40 60 80 100 120

num

ber

d (nm) d (nm)

d (nm)

d (nm)

num

ber

t (nm)

t (nm)

t (nm) t (nm)

Figure 2: Diameter and thickness distributions of the Fe3Se4 nanoplatelets.

7

10 20 30 40 50 60

21

26

2 (º)

counts

(a.

u.)

18

XRD size (nm)

Figure 3: X-ray diffraction patterns of Fe3Se4 nanoplatelets. Continuous (red) linerepresents Rietveld refinement to the XRD data. Average apparent sizes are shown in theright side of the plot.

8

0.0

0.1

0.2

0 10 20 30 40 50 60

-3

-2

-1

0

1

2

3

M (e

mu/

g sam

ple)

dM/d

T (1

0-2 e

mu/

Kg sa

mpl

e)

T (ºC)

250 nm

95 nm193 nm

130 nm

Figure 4: Magnetization of Fe3Se4 nanoplatelets with different average lateraldiameter as a function of temperature and associated derivative. TC is taken asthe minimum of that derivative, i. e., the inflexion point of the M(T) data.

9

15 20 25 3015

20

25

30

35

40

45

T C (º

C)

size XRD (nm)

Figure 5: Average apparent sizes obtained from Rietveld refinement as a functionof the transition temperature TC.

-15 -10 -5 0 5 10 15-3

-2

-1

0

1

2

3

M (e

mu/

g sam

ple)

H (kOe)

T (ºC) -3

57

Figure 6: Dependence of magnetization with the external magnetic field. Curveswere recorded at selected temperatures around room temperature in the Fe3Se4 nanoplateletswith an average diameter of 95 nm.

10

0.0

0.5

1.0

0 10 20 30 40 500

1

2

3

4

Mr (e

mu/

g sam

ple)

HC (k

Oe)

T (ºC)

Figure 7: Remanent magnetization, susceptibility and coercive field of the Fe3Se4

nanoplatelets with average diameter/thickness of 95 nm. All data shown was takenfrom the hysteresis cycles shown in the previous figure.

11

10 100 10000

5

10

15

20

25

30

35

num

ber (

%)

d (nm)

Figure 8: Diameter size distributions of the colloidal suspensions of the Fe3Se4

nanoplatelets. Hydrodynamic size distribution obtained with dynamic light scatter-ing. The average hydrodynamic sizes are compatible with suspensions of isolated Fe3Se4nanoplatelets. In fact, these sizes are systematically smaller than the average diameter ob-tained from electron microscopy and close to the average gyration diameter (Dg) obtainedfor flat cylinders (gyration radius R2

g = (a2/2) + (t2/12) where Rg is half of the gyrationdiameter, a is half of the average diameter and t is the thickness of the nanoplatelets).

12

resetresetresetreset

�

S (field)

�

R (temp.)

�

Q (mag.)

setset resetresetresetset

setresetset set

high H

low H

low M

high M

high T

low T

high H

low H

low M

high M

high T

low T

TC= 36 ºC

set set

0.01

0.1

M (emu/g

sample)

0

200

400

600

800

1000

H (Oe)

0 500 1000 1500 2000 2500 3000

30

35

40

T (ºC)

tim e (s)

0.01

0.1

1

M (emu/g

sample)

0

1000

2000

3000

H (Oe)

0 500 1000 1500

15

20

25

TC= 18 ºCT

C= 22 ºC

T (ºC)

tim e (s)

0 500 100010

15

20

tim e (s)

0

1000

2000

3000

0.01

0.1

1

Figure 9: Logic element representation of the bistable gate (SR latch) and ex-amples of the time diagram of the gate. The gate can be represented by two NORelements in feedback, with two possible inputs, a set (S) and a reset (R), and one output(Q). Example of functioning of the logic gate as a function of time in samples with differentTC , showing the basic operations of the gate based on the response of magnetization (Q) tothe field (S) and temperature (R) inputs.

13

Figure 10: Simulated temperature distribution at equilibrium of the cell cul-ture. Estimation of the temperature distribution on the cell culture chambers containingthe MatrigelTM, nanoplatelets, cells and cell medium when the nanoplatelets are at 42 ◦Cand the total heat flux generated by the nanoplatelets is 0.045 W across a 0.12 cm2 surface.The layered system used in the cell hyperthermia experiments introduces per se a maximumtemperature drop between nanoplatelets and cells of ∼0.5 ◦C and any additional tempera-ture drop will be due to intrinsically lower thermal conductivity at the micro and interfaceswhich would be present even in a non-layered geometry where cells and nanoplatelets weremixed without internalization.

14

1200 1000 800 600 400 200 0 735 730 725 720 715 710 705 700

170 165 160 155 150 65 60 55 50 45

Se 3p

Fe 2p

Se 3d

C 1sO 1s

Intensity (a.u.)

Binding Energy (eV)

130

193

95

250

bulk

Fe 2p

130

193

95

250

bulk

Intensity (a.u.)

Binding Energy (eV)

130

193

95

250

bulk

Intensity (a.u.)

Binding Energy (eV)

Se 3dIntensity (a.u.)

Binding Energy (eV)

130

193

95

250

bulk

Na 1s

Figure 11: XPS data of the Fe3Se4 nanoplatelets (with different average sizes) andreference (bulk) sample The panels show the full spectra and the high resolution Fe 2p,Se 3p and Se 3d spectra. The spectra show small differences among the nanoplatelets withdifferent average sizes and shifts to lower energy values when compared to the microcrys-talline sample. This shift is of the order of the uncertain associated to the C 1s peak usedto correct the effect of the charge neutralizer and therefore it can be an intrinsic effect (as-sociated to different Fe and Se environments/oxidation states) or a spurious effect since theC 1s found in the nanoplatelets is probably different from that found in the microcrystallinesample. Apart this shift, the most noticeable difference between spectra is a satellite peaknear 717.6 eV found only in the microcrystalline sample and the broadening/deconvolutionaround the 725 eV region. One of the samples shows a contamination with Na.

15

Fe 2p

Residual STD = 2.87136Residual STD = 2.03501Residual STD = 1.671Residual STD = 2.89812x 10

2

30

35

40

45

50

55

60

65

70

75

CP

S

735 730 725 720 715 710 705

Binding Energy (eV)

Figure 12: High resolution XPS Fe 2p spectra of the nanoplatelets fitted to that of themicrocrystalline sample, allowing an energy shift.

16

Se 3p

Residual STD = 1.07739Residual STD = 1.68975Residual STD = 1.35885Residual STD = 2.05428

Name

Se 3p

Pos.

163.40

FWHM

1.00

%Area

100.00

Name

Se 3p

Pos.

162.86

FWHM

0.95

%Area

100.00

Name

Se 3p

Pos.

163.73

FWHM

1.00

%Area

100.00

Name

Se 3p

Pos.

163.15

FWHM

0.97

%Area

100.00

Se

3p

Se

3p

Se

3p

Se

3p

x 101

40

50

60

70

80

90

100

CP

S

172 168 164 160 156

Binding Energy (eV)

Figure 13: High resolution XPS Se 3p spectra of the nanoplatelets fitted to that of themicrocrystalline sample, allowing an energy shift.

17

Se 3d

Residual STD = 3.01652Residual STD = 3.30037Residual STD = 3.53702Residual STD = 2.86142

Name

Se 3d

Pos.

57.01

FWHM

1.09

Area

0.08

%Area

100.00

Name

Se 3d

Pos.

56.43

FWHM

0.90

Area

0.07

%Area

100.00

Name

Se 3d

Pos.

56.71

FWHM

0.90

Area

0.05

%Area

100.00

Name

Se 3d

Pos.

56.37

FWHM

0.90

Area

0.08

%Area

100.00

Se

3d

Se

3d

Se

3d

Se

3d

x 101

10

15

20

25

30

35

40

45

50

55

60

CP

S

68 64 60 56 52 48

Binding Energy (eV)

Figure 14: High resolution XPS Se 3d spectra of the nanoplatelets fitted to that of themicrocrystalline sample, allowing an energy shift.

18

10 20 30 40 501E-3

0.01

0.1

1

M (e

mu/

g sam

ple)

T (ºC)

bi-stability temp. range

Figure 15: Standard zero-field cooling (ZFC) and field cooling curves (FC). Mag-netization was recorded as a function of temperature around room temperature at a lowapplied field (Happ = 50 Oe) after cooling from 60 ◦C under zero field (ZFC, open circles)and after cooling under Happ (FC, filled squares). Data was recorded on Fe3Se4 nanoplateletswith average diameter of 95 nm. Below ∼ 35 ◦C the curves diverge and the system can bein a high or a low state. Above that temperature the system can be reset.

19

-18.78

-18.77

-18.76

-18.75

-18.74

-18.73

0.0 0.1 0.2 0.3 0.4

-18.78

-18.77

-18.76

-18.75

-18.74

-18.73

0.0 0.1 0.2 0.3 0.4

570 m W /cm2456 m W /cm

2

342 m W /cm2

before irradiation

after irradiation

field @

fluxgate (

µT)

228 m W /cm2

background field

tim e (m in)

Figure 16: Remanent magnetization Mr measured by the fluxgate magnetometeron the cell wells before and after irradiation with a given power during 5 minutes.The nanoplatelets were set in a high state before irradiation and then approached to thefluxgate magnetometer that senses an increase of magnetic field of the order of 0.05µT dueto dipolar magnetic field created by the Mr of the nanoplatelets (black curves). The curvesshow the ambient background field (grey areas) and the field created by the nanoplateletsplus background (green areas). After irradiation, the nanoplatelets were again approachedto the fluxgate to measure again the dipolar magnetic field (red curves, green area); this fielddecreases as the power increases, vanishing into the background noise for a power between456 and 570 mW/cm2.

20

20 40 60 80 100 120 140 160 180 20099.3

99.4

99.5

99.6

99.7

99.8

99.9

100.0

weight (%

)

T (ºC)

Figure 17: Thermogravimetric analysis on the Fe3Se4 nanoplatelets before phasetransfer. Analysis was performed on a Shimadzu TGA-50 thermogravimetric analyser,using a heating rate of 5 ◦C/min under air atmosphere, with a flow rate of 20 mL/min. Thesample holder was a 5 mm diameter platinum plate and the sample mass was 4.03 mg.

0 5 10 15 20 25 30 35 40 45 50

1E-3

0.01

0.1

0 5 10 15 20 25 30 35 40 45 50

1E-3

0.01

0.1

voltage at fluxgate (V)

tim e (m in)

noise level

low M

high M

voltage at fluxgate (V)

tim e (m in)

noise level

Figure 18: Reproducibility tests. Repetition of set/reset cycles measuring the outputvoltage at the fluxgate magnetometer induced by the high and low magnetization states ofthe logic gates. The high and low states are clearly separated by an order of magnitude ofsignal difference. The oscillations from cycle to cycle are much smaller than this difference.We present 2 groups of measurements performed on different days.

21

0.1 1 10 100 1000 10000

0.08

0.10

0.12

voltage at fluxgate (V)

tim e after set m agnetization high (m in)

Figure 19: Stability tests. Stability over time of the output voltage at the fluxgatemagnetometer after setting magnetization in the high state. The fluctuation of the responseis quite low compared to the difference between the high and low magnetization levels. Along stability means that the logic gates can test a possible temperature threshold surpassoccurring in a wide time frame.

22