betaFeedback

WorksPhysics and Astronomy

Current Paths in an Atomic Precision Advanced Manufactured Device Imaged by Nitrogen-Vacancy Diamond Magnetic Microscopy

Abstract

The recently-developed ability to control phosphorous-doping of silicon at an atomic level using scanning tunneling microscopy (STM), a technique known as atomic-precision-advanced-manufacturing (APAM), has allowed us to tailor electronic devices with atomic precision, and thus has emerged as a way to explore new possibilities in Si electronics. In these applications, critical questions include where current flow is actually occurring in or near APAM structures as well as whether leakage currents are present. In general, detection and mapping of current flow in APAM structures are valuable diagnostic tools to obtain reliable devices in digital-enhanced applications. In this paper, we performed nitrogen-vacancy (NV) wide-field magnetic imaging of stray magnetic fields from surface current densities flowing in an APAM test device over a mm-field of view with μm-resolution. To do this, we integrated a diamond having a surface NV ensemble with the device (patterned in two parallel mm-sized ribbons), then mapped the magnetic field from the DC current injected in the APAM device in a home-built NV wide-field microscope. The 2D magnetic field maps were used to reconstruct the surface current density, allowing us to obtain information on current paths, device failures such as choke points where current flow is impeded, and current leakages outside the APAM-defined P-doped regions. Analysis on the current density reconstructed map showed a projected sensitivity of ~0.03 A/m, corresponding to a smallest detectable current in the 200 μm-wide APAM ribbon of ~6 μA. These results demonstrate the failure analysis capability of NV wide-field magnetometry for APAM materials, opening the possibility to investigate other cutting-edge microelectronic devices.

Introduction

Atomic-precision control over crystalline Si synthesis and processing are putting pure quantum mechanics phenomena (e.g. spin and tunnel effects) on the Si device engineer’s palette of reliable options for creating new microelectronics. One recent addition[1, 2] to the palette is a technique known as atomic-precision advanced manufacturing (APAM). A typical APAM process first creates a lithographic pattern of Si dangling bonds on a hydrogen-passivated crystalline Si surface, using scanning tunneling microscopy (STM). Next, the Si surface is exposed to phosphine gas that selectively adsorbs on sites where Si dangling bonds have been exposed, leading to atomically-precise planar structures made of P-donors[3, 4]. APAM allows us to dope Si in full 2D atomic-precision[5, 6, 7, 8, 9], to form structures such as wires[5, 10], quantum dots[11], and few- and single-dopant features[12, 13] with lattice-site (3.8 Å) control[9]. Doped features can be positioned, arrayed, coupled, and gated with atomic precision to form tunnel junctions[14], single-atom transistors[12], and a host of other possibilities, all compatible with near-standard CMOS manufacturing[6, 7, 15]. This gives us an exciting tool to explore far-reaching ideas, such as quantum computing[8], quantum simulation[16], and digital electronics that incorporate controlled quantum effects into Si foundry-compatible platforms[6, 8, 17, 18].

Currently, APAM devices work only in cryogenic conditions, hence their use is limited to applications where low temperatures are required (e.g. in quantum computing)

ing) to achieve long coherence times. Realizing broader possible applications for APAM techniques in room-temperature conventional electronics is a big challenge, as it requires understanding of carrier transport in ultrathin high-density P-doped Si structures. Electronic transport in APAM-doped materials and structures has been investigated and understood at cryogenic conditions (T<4.2T < 4.2 K)[19]. APAM-doped regions would dominate carrier transport just by virtue of the high APAM doping levels (102210^{22} cm−3^{-3}), while doping in surrounding Si was kept sufficiently low (≪3⋅1018\ll3 \cdot10^{18} cm−3^{-3}, i.e. the metal-insulator transition) that leakage paths through the substrate and cap layers are frozen out at 4 K, contributing little to a device’s transport properties. In higher-TT applications, where substrate freeze-out does not occur, unintended leakage currents can be much more than a nuisance background power-drain, and can potentially swamp the intended current channels and mask the intended device functionality[17]. Thus, understanding and engineering charge carrier confinement to intended doped transport pathways is an integral challenge particularly relevant for room-temperature quantum-enhanced digital electronics. As a result, methods to detect and map the carrier flow in APAM-doped structures are valuable tools to build the understanding necessary to achieve reliable devices. More generally, a broader suite of failure analysis tools validated not only for APAM, but for a wide range of novel materials, is timely and relevant[20, 21, 22, 23, 24, 25].

In this work, we used an ensemble of nitrogen-vacancy (NV) centers in diamond to map the magnetic fields generated erated by currents injected in an APAM device. For a proof-of-concept demonstration of NV magnetometry as an APAM materials diagnostic tool, we investigated a device made of straightforward micron-scale ribbon-shaped structures defined by mesas etched from blanket P-delta-doped Si. These structures are recognized, and utilized heavily, throughout the APAM community as a valid witness material for APAM doping processes[26, 27]. NV magnetometry[28, 29, 30, 31] non-invasively measures stray magnetic fields by exploiting the magnetically-sensitive NV electronic spin ground-state, the ability to manipulate the sublevel populations using a resonant microwave field, and the optical spin readout by spin-dependent photoluminescence (PL). By collecting the NV fluorescence on a CMOS camera in a home-built optical microscope, we obtained a 2D magnetic field map over a few-millimeter field of view with micrometer-scale spatial resolution[32, 33]. We then reconstructed the current paths within the device from the measured magnetic images, revealing current flow features of a functioning APAM device, as well as that of a faulty one due to micrometer- and nanometer-scale material damage. This approach extends the suite of existing magnetic imaging tools used in the electronics failure analysis community (e.g. scanning gate microscopy[34], scanning SQUID microscopy[35], and magnetic force microscopy[36]), with the added benefits of high spatial resolution, high instrument reliability, wide-field data collection, and non-invasive ambient-conditions operation.

Materials and Methods

Experimental Setup

To perform wide-field magnetic imaging, we used a home-built fluorescence microscope (schematic shown in Figure 1(a))[33]. The diamond sensor, placed on top of the APAM device, was illuminated from the side with a 532 nm laser beam, with a laser power of 1.1 W. The laser beam was expanded with a 400 mm focal length lens in order to obtain a uniform illumination in the camera field of view (∼2.4×3.8\sim2.4 \times3.8 mm2^2). The NV fluorescence was collected by a 5x objective (0.13 numerical aperture) and filtered by a long-pass filter (650 nm cutoff wavelength) before being imaged through a tube lens (f=100f = 100 mm) onto a CMOS camera. The camera exposure time was set to 25 ms and a 4×44 \times4 pixel binning was applied to obtain an image of 250×400 pixels with a pixel size of 9.5 μ\mum. The microscope was placed inside a 3-axis Helmholtz coil set, used to apply a static B0≃1B_0 \simeq1 mT bias magnetic field along the NV axis, defined as the [111] diamond crystallographic direction as shown in Figure 1(b) and (c). The bias magnetic field lifts the degeneracy between the NV ground-state spin sublevels ms=±1m_s = \pm1 required to perform optically-detected magnetic resonance (ODMR) spectroscopy. The resonance frequencies of the [111] oriented NVs, as shown in Figure 1(d), are shifted from the zero-field splitting f0≃2.87f_0 \simeq2.87 GHz by ±γB111\pm\gamma B_{111}, with γ≃28\gamma\simeq28 kHz μ\muT−1^{-1} being the NV gyromagnetic ratio and B111B_{111} the magnetic field along the NV axis. As the laser pumps NVs into the mS=0m_S = 0 bright sublevel, we performed ODMR spectroscopy by applying a microwave (MW) field to drive transitions from the mS=0m_S = 0 to the mS=±1m_S = \pm1 dark sublevels while monitoring the NV PL emission: when the MW is on resonance with the mS=±1m_S = \pm1, the PL intensity is reduced. In our apparatus, a copper loop placed above the diamond provided the MW excitation. The MW signal was generated by a TPI1002-A MW source, pulsed by a Mini-Circuits ZASWA2-50DRA+, then amplified into the copper loop with a Mini-Circuits ZHL-16W-43-S+.

Image not extracted

Figure 1. (a) Schematics of the NV wide-field magnetic microscope. (b) NV crystal structure, with the [111] direction indicated by the red arrow. (c) NV center [111] direction orientation in the reference frame set by the three Helmholtz coil axis. (d) NV center ground-state spin sublevels energy diagram, showing the zero-field splitting f0≃2.87f_0 \simeq2.87 GHz and the Zeeman effect lifting the ms=±1m_s = \pm1 degeneracy. (e) Diamond-device integration method.

NV Diamond Sample

Our NV diamond sensor was a 4×4×0.5 mm34 \times4 \times0.5\ \mathrm{mm}^{3} electronic grade diamond (native nitrogen density <5< 5 ppb), that was CVD overgrown with a 4 µm-thick 12C{}^{12}\mathrm{C}-enriched diamond layer doped with 25 ppm of 14N{}^{14}\mathrm{N} on one surface (process performed by Applied Diamond, Inc). After overgrowth, it was irradiated with a 1 MeV energy, 1.2×1018 cm−21.2 \times10^{18}\ \mathrm{cm}^{-2} fluence electron beam and vacuum-annealed as in Ref.[37] to form a NV ensemble. Finally, the diamond was cleaned with triacid solution (1:1:1 sulfuric, nitric and perchloric acids) at 250 °C for 1 hour. To avoid photocarriers excitation in the imaged device, we glued the non-NV-containing diamond side to a glass coverslip with a UV-curing transparent glue. Both the diamond and the

coverslip were then coated with a three-layer stack made of 5 nm of Ti (to improve adhesion), 150 nm of Ag (to reflect photons and prevent photocurrent excitation), and 150 nm of Al2_2O3_3 (to avoid contact shorting in the device)[38]. As shown in Figure 1(e), our measurements were performed with the diamond placed on top of the device, with the NV surface facing down. As a demonstration of the importance of protecting the device under studying, results of magnetic imaging without the three-layer stack and the consequences on APAM device functioning are reported in the supplementary materials[39].

Characterization Techniques

SEM analysis was performed with a FEI Nanolab 650 microscope, operated in secondary electron mode with an acceleration voltage of 10 keV and at a working distance of 5 mm. Laser profilometry was done by using a Keyence VK-X150 microscope equipped with a 658 nm laser.

APAM Device

The schematic of the APAM device fabrication process is shown in Figure 2(a). APAM phosphorus delta-doped layer devices were prepared on flash-cleaned float zone Si(100) substrates through UHV dosing with phos phine and subsequent encapsulation with 30 nm of low temperature epitaxial silicon. The samples were then processed using standard cleanroom procedures using an all-optical lithography approach. Two Hall bar structures were etched into the silicon substrate through a Bosch etch, taking care to etch deep enough (10 μm\mu\mathrm{m}) to isolate the delta layer from the surrounding silicon substrate. An ICP silicon etch was then used to etch several shallow vias through the APAM silicon cap to the APAM delta layer. Finally, electrical contact to the APAM delta layer was achieved by the addition of 300nm-thick Ti/Al bond pads using a standard metals liftoff process. More details on the APAM device fabrication process are reported in Ref. 6. Figure 2(b) shows the schematic top view of the chip, whereas the final APAM device optical picture is shown in figure 2(c).

Image not extracted

Figure 2. (a) APAM device fabrication process flow, shown in cross section. (b) Schematic view of the top side of the chip, showing the etched delta-doped layer having the shape of two parallel 200 µm wide ribbons (dark green) and contact metallization (light green). (c) Final APAM device optical image.

Results and Discussion

Magnetic Field Measurements

We injected a current of I=150 μAI=150\,\mu\mathrm{A} (applied voltage of 1.9 V) through one of the APAM device ribbons, as shown in figure 3(a). We performed wide-field ODMR spectroscopy, collecting emitted NV PL while sweeping the MW frequency, obtaining an ODMR spectrum for each camera pixel. We then fit the ODMR spectra for each pixel, shown in figure 3(b), to determine the resonance frequencies. The frequency difference between the two resonance frequencies Δf\Delta f is used to calculate the value of the magnetic field along the NV axis B111=Δf/2γB_{111}=\Delta f/2\gamma for each pixel, allowing us to obtain a 2D magnetic field map. To isolate the contributions of the injected current to the magnetic field (and remove any background magnetic fields), we collected two wide-field magnetic image measurements, one with positive current II, the other with a current of opposite sign −I-I. To improve the signal-to-noise ratio, each measurement was repeated and averaged 10 times for a total acquisition time of ∼3\sim3 h. The normalized difference of the two images B111(I)=(B(+I)−B(−I))/2B_{111}(I)=(B(+I)-B(-I))/2 was then calculated to remove the background magnetic field. The B111(I)B_{111}(I) magnetic field map resulting from wide-field ODMR spectroscopy is displayed in figure 3(c).

Stand-off Distance Estimation

To estimate the stand-off distance hh between the NV-layer and the APAM device, we used the following equations for the magnetic field generated along a line-cut of an infinitely long 2D ribbon having width w=200w = 200 µm and carrying a current II[40]:

By=μ0I2πw[arctan⁡(y−w/2h)−arctan⁡(y+w/2h)],B_y = \frac{\mu_0 I}{2\pi w}\left[\arctan\left(\frac{y-w/2}{h}\right)-\arctan\left(\frac{y+w/2}{h}\right)\right],
Bz=μ0I4πwlog⁡(h2+(y−w/2)2h2+(y+w/2)2).(1)B_z = \frac{\mu_0 I}{4\pi w}\log\left(\frac{h^2+(y-w/2)^2}{h^2+(y+w/2)^2}\right). \tag*{(1)}

Here yy and zz are respectively the directions along the width of the ribbons and perpendicular to the device plane and μ0=4π×10−7\mu_0 = 4\pi\times10^{-7} m·T/A. The magnetic field projection along the NV axis is given by:

B111=Bysin⁡ϕsin⁡θ+Bzcos⁡θ,(2)B_{111} = B_y \sin\phi\sin\theta+ B_z\cos\theta, \tag*{(2)}

with ϕ≃89∘\phi\simeq89^\circ and θ≃57∘\theta\simeq57^\circ being the two angles defining the NV-axis direction. θ\theta and ϕ\phi are respectively the angle from zz and the angle in the xyxy plane, defining the direction of the bias magnetic field aligned with the [111] diamond axis, as shown in Figure 1(c).

Finally, the finite NV layer thickness tNV=4t_{\mathrm{NV}} = 4 µm is taken into account by replacing hh with zz, and averaging over the interval h≤z≤h+tNVh \le z \le h+t_{\mathrm{NV}}[41]:

BNV=1tNV∫hh+tNVB111(y,z) dz,(3)B_{\mathrm{NV}} = \frac{1}{t_{\mathrm{NV}}}\int_h^{h+t_{\mathrm{NV}}} B_{111}(y,z)\,dz, \tag*{(3)}

resulting in h=(14±2)h = (14 \pm2) µm, as shown in Figure 3(d). The error bars on the experimental data in Figure 3(d) come from the δB\delta B standard deviation (noise floor) of the magnetic fields measured over a line-cut away from the ribbon, i.e. where ⟨B111⟩=0\langle B_{111}\rangle= 0, resulting in δB=0.04\delta B = 0.04 µT.

Image not extracted

Figure 3. (a) Optical image of the APAM device showing over which ribbon terminals a 150 μA\mu\mathrm{A} current was injected. (b) Typical ODMR spectrum for a single pixel for a positive (red dots) and negative (green dots) injected current. Each resonant transition is split into three peaks due to the 14N{}^{14}\mathrm{N} hyperfine interaction. The spectra are fit with six Lorentzian lineshapes (solid lines), yielding the value of B111B_{111} for each pixel. (c) NV magnetic field map for the same area shown in (a) obtained by (B(+I)−B(−I))/2(B(+I)-B(-I))/2. (d) Magnetic field profile along the white dashed line in (c) fit with Eq. (3) (solid line) resulting in a stand-off distance of h=(14±2) μmh=(14\pm2)\,\mu\mathrm{m}.

Current Reconstruction

After finding the stand-off distance, we reconstructed the surface current density K(x,y)K(x,y) from the single vector component magnetic field map B111B_{111}. We assumed that the currents flowing in the APAM device are quasistatic and confined in a 2D sheet, and we used the inverted Biot-Savart law in Fourier space to obtain[42, 43, 44]:

K^x(kx,ky)=w(k,λ)kyg(h,kz)[eyky−exkx+iezk]b(kx,ky,h),\hat{K}_x(k_x,k_y) = \frac{w(k,\lambda)k_y}{g(h,k_z)[e_yk_y-e_xk_x+ie_zk]}b(k_x,k_y,h),
K^y(kx,ky)=w(k,λ)kxg(h,kz)[exkx−eyky−iezk]b(kx,ky,h).(4)\hat{K}_y(k_x,k_y) = \frac{w(k,\lambda)k_x}{g(h,k_z)[e_xk_x-e_yk_y-ie_zk]}b(k_x,k_y,h). \tag*{(4)}

Here {K^x,K^y}\{\hat{K}_x,\hat{K}_y\} and b(kx,ky,h)b(k_x,k_y,h) are respectively the current density components and the magnetic field map in Fourier space, kxk_x and kyk_y are the Fourier-space wave vector components, k=(kx2+ky2)1/2k = (k_x^2+k_y^2)^{1/2} and ${e_x,e_y,e_z} = {sin⁡ϕsin⁡θ,cos⁡ϕsin⁡θ,cos⁡θ}\{\sin\phi\sin\theta, \cos\phi\sin\theta, \cos\theta\} is the NV axis orientation. The function g(k,z)g(k,z) is the Green’s function

g(k,z)=μ02e−kz,(5)g(k,z) = \frac{\mu_0}{2}e^{-kz}, \tag*{(5)}

which requires a window function w(k,λ)w(k,\lambda) to suppress the measurement white noise at high spatial frequencies kk when g(k,z)g(k,z) tends to zero. To achieve this, we use a Hann window function

w(k,kmax⁡)={12[1+cos⁡(πk/kmax⁡)]if ∣k∣<kmax⁡0otherwise(6)w(k,k_{\max}) = \begin{cases} \frac{1}{2}[1+\cos(\pi k/k_{\max})] & \text{if } |k|<k_{\max} \\ 0 & \text{otherwise} \end{cases} \tag*{(6)}

with cutoff wave vector kmax⁡k_{\max}. Noise rejection of high kk values reduces the spatial resolution of the calculated current distribution (which is already limited by the stand-off distance), so kmax⁡k_{\max} is found empirically to adjust the trade-off between these two effects. Finally, the inverse Fourier transform of Eq. (4) leads to the current density components KxK_x and KyK_y as well as the current density magnitude ∣K⃗∣=(Kx2+Ky2)1/2|\vec{K}|=(K_x^2+K_y^2)^{1/2}.

Functioning APAM Device

The reconstructed current density for the magnetic field map of Figure 4(a) is shown in Figure 4(b,Figure 4c,Figure 4d). To check the quality of the current reconstruction algorithm, we integrated the current density over the width of the ribbon to obtain a total current of IR=152I_R=152 µA (Figure 4(e)). The value of the surface current density in the ribbon KRK_R as function of xx is shown in Figure 4(f), resulting in a constant value of KR=(0.80±0.03)K_R=(0.80\pm0.03) A/m. The resulting current obtained by integrating KRK_R over the width of the ribbon is shown in Figure 4(g). The constant value of IR=(150±3)I_R=(150\pm3) µA proves the current stability across the P-doped region and matches the nominal injected current of 150 µA. The background surface current density KBGK_{BG} measured in a region away from the ribbon is also shown in Figure 4(f), showing a mean offset baseline of KBG=0.05K_{BG}=0.05 A/m and a standard deviation of δK=0.03\delta K=0.03 A/m. The DC offset KBGK_{BG} in the KK map is an artifact of the current reconstruction algorithm arising when KK is calculated from a single axis magnetic field measurement[44] while the standard deviation δ=0.03\delta=0.03 A/m represents the noise floor. δK\delta K gives a measurement of the surface current density sensitivity, leading to a smallest detectable current (i.e. current measured with a SNR = 1) in the 200 µm wide ribbon of ∼6\sim6 µA.

Image not extracted

Figure 4. (a) Magnetic field map for an injected current of 150 µA. (b) Reconstructed current density magnitude ∣K~∣=(Kx2+Ky2)−1/2|\tilde{K}| = (K_x^2 + K_y^2)^{-1/2}. (c)-(d) KxK_x and KyK_y component respectively. The scale bar is the same for (a-d). (e) Current density along the white dashed line in (b), leading to an integrated current flowing in the ribbon of IR=152I_R = 152 µA. (f) Surface current density as function of xx measured on a 1D line-cut in the center of the ribbon (blue dots) and in the background (green dots). The KRK_R and KBGK_{BG} values are given by the average (solid lines) while the uncertainty is the standard deviation (shaded areas). (g) Current as function of xx inside the ribbon (blue dots) obtained by integrating ∣K~∣|\tilde{K}| over the width of the ribbon. The IRI_R values is given by the average (solid line) while the uncertainty is the standard deviation (shaded area).

Faulty APAM Device

We can also detect failures in our APAM devices. An example is shown in Figure 5(a) for an injected current of 150 µA (applied voltage of ∼18\sim18 V), where there are sections of the ribbon for which we can observe impeded current flow. Figure 5(b) shows a zoom of the current path irregularities: the current direction around holes and choke points are indicated by the arrows, which have lengths proportional to the local current density magnitude. To assess the origins of these irregularities in the current flow, we compared the measured current density maps with optical images of the same areas (Figure 5(c)), as well as SEM (Figure 5(d)) and laser profilometry (Figure 5(e)) analysis. From the set of optical images in Figure 5(c) we could observe a correlation between current flow irregularities and material damage on the APAM device surface. SEM analysis on the left image of Figure 5(d) shows the presence of two pairs of scratches each with lateral size of ≃250\simeq250 nm, as shown in right panel of Figure 5(d). We then used laser profilometry to obtain a 3D altitude image of the same areas. As seen in Figure 5(e) some of the cracks on the device surface are several microns deep, enough to have removed the conductive P-doped region below Si surface. This allows us to attribute the current flow irregularities to the presence of material damage, such as cracks and scrapes. The overlay between current density maps and optical images, shown in Figure 5(f), helps to see where the current is flowing around the device defects.

Image not extracted

Figure 5. (a) Reconstructed current density magnitude ∣K⃗∣|\vec{K}| of a damaged ribbon for an injected current of 150 µA. (b) Zoom of areas of current flow irregularities. The arrows point in the direction of K⃗\vec{K} with length proportional to ∣K⃗∣|\vec{K}|. (c) Optical images corresponding to the same regions in (b). (d) SEM images with different magnification of the area in (c)-left, showing four scratches in the APAM device surface highlighted by the black arrows and a zoom of one of the scratches having a lateral size of ∼250 nm. (e) Laser profilometry analysis of the same regions in the center and right images of (c). (f) Overlay of optical and ∣K⃗∣|\vec{K}| images. The scale bars for (b), (c), (e) and (f) are the same.

Current Leakage Between APAM structures

Finally, with the aim of detecting current leakage paths, we intentionally created leakages between the P-doped ribbons by connecting both of them to the current source, as shown in Figure 6(a). By doing this, we expect the current to leave the bottom ribbon and flow through the Si substrate to reach the top one. Unsurprisingly, to inject a current of I=175I=175 µA we need to apply a voltage of ∼70\sim70 V, ∼10\sim10× higher than the previous cases, consistent with the current flowing through an undoped path. The resulting magnetic field map and reconstructed current density, reported in Figure 6(b) and (c) respectively, show that the current is flowing in both ribbons. For a quantitative analysis we integrated the current density over the ribbons to calculate the current flowing along xx, with the result shown in Figure 6(d). The error bars on the integrated current are calculated starting from the uncertainty on the measured magnetic field δB=0.04\delta B=0.04 µT, that is propagated through the current reconstruction algorithm to obtain δI=9\delta I=9 µA. We observed that the current profiles are consistent with the current being injected in the bottom ribbon and collected from the top one. Interestingly, the sum of the currents flowing in the two ribbons matches the total injected current only close to the injection and collection points, implying that there are leakage currents in between the ribbons that we do not detect. More quantitatively, the unobservable current, i.e. leakage current defined as the difference between the nominal injected current and total measured current (black curve in Figure 6(d)), is at most 39±1339\pm13 µA. As the measured surface current density sensitivity is δK=0.03\delta K = 0.03 A/m (see Figure 4(f)), the leak current to be detected with a SNR = 1 must flow in a sheet less than ∼1 mm wide. We postulate that we could not directly observe the current leakage path for several reasons. The first explanation is that the leakage path is three dimensional, making the current density reconstruction unreliable due to the 2D current density assumption of our model. The second explanation is that the leakage currents, no longer confined in the 2D P-doped regions, are spread over a wide area, resulting in a surface current density smaller than δK\delta K. Finally, if the leakage current behaves as a uniform sheet between the two ribbons, then its field would be largely along the xx direction (except at the edges), which is orthogonal to the NV sensing direction and would be largely invisible. Although the field from such a leakage current would be too weak to detect with the present apparatus, a follow-up experiment with an improved δK\delta K that measures the vector magnetic fields for each pixel may be able to detect it.

Image not extracted

Figure 6. (a) I=175I = 175 µA current injection configuration. Dashed white box is the region imaged in (b) and (c). (b) Measured magnetic field map. (c) Reconstructed current density ∣K~∣|\widetilde{K}|. The scale bar for (b) and (c) is the same. (d) Integrated current density for the top (red curve) and bottom ribbon (blue curve), with the sum between the two (black curve). The nominal injected current (purple curve) and the background (average value solid green line, standard deviation shaded green area) are also reported.

Conclusion

In summary, we have shown the capability to map the magnetic fields in an APAM device with micron-scale resolution over a millimeter-scale field of view area using NV wide-field magnetic imaging under ambient conditions without interfering with the operation of the device under study (see supplementary materials[39]). From the measured magnetic field maps, we reconstructed the surface current densities, leading to wide-field imaging of the current flowing in the device. That allowed us to obtain information on the APAM device properties and detect unexpected behavior, such as current path irregularities and leakages. Current density sensitivity is of order 0.03 A/m, leading to a minimum detectable current of ∼6 µA in the 200 µm wide APAM ribbon. Further work will upgrade this experiment to detect and reconstruct 3D currents[45] as well as AC signals in the MHz or GHz range with appropriate NV manipulation techniques[46, 47]. The development of a NV-based diagnostic tool able to image where current is flowing and detect failures or leakages down to 0.03 A/m could benefit the whole microelectronics community, with application not limited to APAM devices but extended to CMOS based technologies[48].

Acknowledgements

We thank George Burns, Michael Titze, and Edward Bielejec (Ion Beam Laboratory, Sandia National Laboratories) for help with diamond preparation. Sandia National Laboratories is a multi-mission laboratory managed and operated by National Technology and Engineering Solutions of Sandia, LLC, a wholly owned subsidiary of Honeywell International, Inc., for the DOE’s National Nuclear Security Administration under contract DE-NA0003525. This work was funded, in part, by the Laboratory Directed Research and Development Program and performed, in part, at the Center for Integrated Nanotechnologies, an Office of Science User Facility operated for the U.S. Department of Energy (DOE) Office of Science. This paper describes objective technical results and analysis. Any subjective views or opinions that might be expressed in the paper do not necessarily represent the views of the U.S. Department of Energy or the United States Government.

Supplemental material for Current Paths in an Atomic Precision Advanced Manufactured Device Imaged by Nitrogen-Vacancy Diamond Magnetic Microscopy

Luca Basso; Pauli Kehayias; Jacob Henshaw; Maziar Saleh Ziabari; Heejun Byeon; Michael P. Lilly; Ezra Bussmann; Deanna M. Campbell; Shashank Misra; Andrew M. Mounce

Diamond sensor reflective coating

To avoid creating photoexcited carriers in the APAM device from the NV pump laser impinging on the Si substrate, we deposited on the NV-center enriched diamond surface a three-layer stack reflective coating (see section 2.2 in the main text). To highlight the importance of the protective coating, we performed the same magnetic field measurement with the uncoated diamond. Figure S1 shows a typical magnetic field and reconstructed current maps obtained with the uncoated diamond when the current is injected only in the top ribbon, as shown in Figure S1. Despite injecting the current in only one ribbon, we could clearly also observe the current flowing in the unplugged bottom ribbon, both in the magnetic field (Figure S1) and in the reconstructed surface current density maps (Figure S1-(Figure S1)). This is attributed to laser going through the diamond and impinging on the below APAM device Si surface, leading to free-carrier excitation and a consequent presence of photocurrents in the device. For comparison, Figure 3(c) in the main text does not show any current leaving the ribbon and flowing in the unplugged ribbon when the coated diamond is used for the experiment. The presence of photocurrents is also confirmed by simultaneously measuring the resistance during magnetic field imaging. Across the top ribbon, as in Figure S1, the resistance drops from R≈120R \approx120 kΩ\Omega to R≈16R \approx16 kΩ\Omega when the laser is turned on and the diamond is uncoated, whereas no relevant change in RR is observed when the diamond is coated.

Image not extracted

Figure S1. Magnetic field measurement without diamond protective coating. (a) APAM device optical picture showing over which ribbon terminals a 100 μ\muA current is injected. (b) Measured magnetic field map. (c) Reconstructed surface current density magnitude ∣K~∣=(Kx2+Ky2)1/2|\tilde{K}| = (K_x^2 + K_y^2)^{1/2}. (d)-(e) KxK_x and KyK_y components respectively. The scale bar is the same for (a)-(e).

References

  1. [1]F. J. Rueß, L. Oberbeck, M. Y. Simmons, K. E. J. Goh, A. R. Hamilton, T. Hallam, B. R. Schofield, N. J. Curson, and R. G. Clark, “Toward atomic-scale device fabrication in silicon using scanning probe microscopy,” Nano Lett. 4, 1969–1973 (2004).DOI
  2. [2]T.-C. Shen, J. Lin, T. Schenkel, S. Robinson, J.-Y. Ji, C. Yang, R.-R. Du, and J. Tucker, “Nanoscale electronics based on two-dimensional dopant patterns in silicon,” J. Vac. Sci. Technol. B 22, 3182–3185 (2004).DOI
  3. [3]M. Simmons, F. Ruess, K. Goh, T. Hallam, S. Schofield, L. Oberbeck, N. Curson, A. Hamilton, M. Butcher, R. Clark, et al., “Scanning probe microscopy for silicon device fabrication,” Mol. Simul. 31, 505–515 (2005).DOI
  4. [4]F. J. Rueß, W. Pok, T. C. Reusch, M. J. Butcher, K. E. J. Goh, L. Oberbeck, G. Scappucci, A. R. Hamilton, and M. Y. Simmons, “Realization of atomically controlled dopant devices in silicon,” Small 3, 563–567 (2007).
  5. [5]S. McKibbin, G. Scappucci, W. Pok, and M. Simmons, “Epitaxial top-gated atomic-scale silicon wire in a three-dimensional architecture,” Nanotechnology 24, 045303 (2013).DOI
  6. [6]D. R. Ward, M. T. Marshall, D. Campbell, T.-M. Lu, J. C. Koepke, D. A. Scrygmeour, E. Bussmann, and S. Misra, “All-optical lithography process for contacting nanometer precision donor devices,” Appl. Phys. Lett. 111, 193101 (2017).
  7. [7]T. Škereň, N. Pascher, A. Garnier, P. Reynaud, E. Rolland, A. Thuaire, D. Widmer, X. Jehl, and A. Fuhrer, “Cmos platform for atomic-scale device fabrication,” Nanotechnology 29, 435302 (2018).DOI
  8. [8]D. R. Ward, S. W. Schmucker, E. M. Anderson, E. Bussmann, L. Tracy, T.-M. Lu, L. N. Maurer, A. Baczewski, D. M. Campbell, M. T. Marshall, and S. Misra, “Atomic precision advanced manufacturing for digital electronics,” Electronic Device Failure Analysis 22, 4–11 (2020).DOI
  9. [9]E. Bussmann, R. E. Butera, J. H. Owen, J. N. Randall, S. M. Rinaldi, A. D. Baczewski, and S. Misra, “Atomic-precision advanced manufacturing for si quantum computing,” MRS Bulletin 46, 607–615 (2021).DOI
  10. [10]B. Weber, S. Mahapatra, H. Ryu, S. Lee, A. Fuhrer, T. Reusch, D. Thompson, W. Lee, G. Klimeck, L. C. Hollenberg, et al., “Ohm’s law survives to the atomic scale,” Science 335, 64–67 (2012).DOI
  11. [11]A. Fuhrer, M. Fuchsle, T. Reusch, B. Weber, and M. Simmons, “Atomic-scale, all epitaxial in-plane gated donor quantum dot in silicon,” Nano Lett. 9, 707–710 (2009).DOI
  12. [12]M. Fuechsle, J. A. Miwa, S. Mahapatra, H. Ryu, S. Lee, O. Warschkow, L. C. Hollenberg, G. Klimeck, and M. Y. Simmons, “A single-atom transistor,” Nat. Nanotechnol. 7, 242–246 (2012).
  13. [13]Y. He, S. Gorman, D. Keith, L. Kranz, J. Keizer, and M. Simmons, “A two-qubit gate between phosphorus donor electrons in silicon,” Nature 571, 371–375 (2019).DOI
  14. [14]X. Wang, J. Wyrick, R. V. Kashid, P. Namboodiri, S. W. Schmucker, A. Murphy, M. D. Stewart, and R. M. Silver, “Atomic-scale control of tunneling in donor-based devices,” Commun. Phys. 3, 1–9 (2020).DOI
  15. [15]T. Skereň, S. A. Köster, D. Douhard, C. Fleischmann, and A. Fuhrer, “Bipolar device fabrication using a scanning tunnelling microscope,” Nature Electronics 1, 1–7 (2020).DOI
  16. [16]X. Wang, E. Khatami, F. Fei, J. Wyrick, P. Namboodiri, R. Kashid, A. F. Rigosi, G. Bryant, and R. Silver, “Quantum simulation of an extended fermi-hubbard model using a 2d lattice of dopant-based quantum dots,” arXiv preprint arXiv:2110.08982 (2021).DOI
  17. [17]X. Gao, L. A. Tracy, E. M. Anderson, D. M. Campbell, J. A. Ivie, T.-M. Lu, D. Mamaluy, S. W. Schmucker, and S. Misra, “Modeling assisted room temperature operation of atomic precision advanced manufacturing devices,” 2020 International Conference on Simulation of Semiconductor Processes and Devices (SISPAD), 277–280 (2020).DOI
  18. [18]T.-M. Lu, X. Gao, E. M. Anderson, J. P. Mendez, D. M. Campbell, J. A. Ivie, S. W. Schmucker, A. Grine, P. Lu, L. A. Tracy, R. Arghavani, and S. Misra, “Path towards a vertical fet enabled by atomic precision advanced manufacturing,” 2021 Silicon Nanoelectronics Workshop (SNW), 1–2 (2021).
  19. [19]D. Mamaluy, J. P. Mendez, X. Gao, and S. Misra, “Revealing quantum effects in highly conductive δ-layer systems,” Commun. Phys. 4 (2021).DOI
  20. [20]E. Bussmann, M. Rudolph, G. Subramania, S. Misra, S. Carr, E. Langlois, J. Dominguez, T. Plyum, M. Lilly, and M. Carroll, “Scanning capacitance microscopy registration of buried atomic-precision donor devices,” Nanotechnology 26, 085701 (2015).DOI
  21. [21]D. Scrymeour, A. Baca, K. Fishgrab, R. Simonson, M. Marshall, E. Bussmann, C. Nakakura, M. Anderson, and S. Misra, “Determining the resolution of scanning microwave impedance microscopy using atomic-precision buried donor structures,” Appl. Surf. Sci. 423, 1097–1102 (2017).
  22. [22]G. Gramse, A. Kölker, T. Lim, T. J. Stock, H. Solanki, S. R. Schofield, E. Brinciotti, G. Aeppli, F. Kienberger, and N. J. Curson, “Nondestructive imaging of atomically thin nanostructures buried in silicon,” Science Advances 3, e1602586 (2017).DOI
  23. [23]A. M. Katzenmeyer, T. S. Luk, E. Bussmann, S. Young, E. M. Anderson, M. T. Marshall, J. A. Ohlhausen, P. Kotula, P. Lu, D. M. Campbell, et al., “Assessing atomically thin delta-doping of silicon using mid-infrared ellipsometry,” J. Mater. Res. 35, 2098–2105 (2020).DOI
  24. [24]G. Gramse, A. Koelker, T. Skeren, T. J. Stock, G. Aeppli, F. Kienberger, A. Fuhrer, and N. J. Curson, “Nanoscale imaging of mobile carriers and trapped charges in delta doped silicon p–n junctions,” Nat. Electron. 3, 531–538 (2020).DOI
  25. [25]A. Kölker, G. Gramse, T. J. Stock, G. Aeppli, and N. J. Curson, “In operando charge transport imaging of atomically thin dopant nanostructures in silicon,” Nanoscale 14, 6437–6448 (2022).
  26. [26]L. Oberbeck, N. Curson, M. Simmons, R. Brenner, A. Hamilton, S. Schofield, and R. Clark, “Encapsulation of phosphorus dopants in silicon for the fabrication of a quantum computer,” Appl. Phys. Lett. 81, 3197–3199 (2002).DOI
  27. [27]J. A. Hagmann, X. Wang, P. Namboodiri, J. Wyrick, R. Murray, M. Stewart Jr, R. M. Silver, and C. A. Richter, “High resolution thickness measurements of ultrathin si: P monolayers using weak localization,” Appl. Phys. Lett. 112, 043102 (2018).DOI
  28. [28]M. W. Doherty, N. B. Manson, P. Delaney, F. Jelezko, J. Wrachtrup, and L. Hollenberg, “The nitrogen-vacancy colour centre in diamond,” Physics Reports 528, 1–45 (2013).DOI
  29. [29]L. Rondin, J.-P. Tetienne, T. Hingant, J.-F. Roch, P. Maletinsky, and V. Jacques, “Magnetometry with nitrogen-vacancy defects in diamond,” Rep. Prog. Phys. 77, 056503 (2014).
  30. [30]M. Chipaux, A. Tallaire, J. Achard, S. Pezzagna, J. Meijer, V. Jacques, J.-F. Roch, and T. Debuisschert, “Magnetic imaging with an ensemble of nitrogen-vacancy centers in diamond,” Eur. Phys. J. D 69, 166 (2015).DOI
  31. [31]P. Kehayias, E. Bussmann, T. M. Lu, and A. M. Mounce, “A physically unclonable function using nv diamond magnetometry and micromagnet arrays,” J. Appl. Phys. 127, 203904 (2020).DOI
  32. [32]D. R. Glenn, R. R. Fu, P. Kehayias, D. Le Sage, A. Lima, B. P. Weiss, and R. L. Walsworth, “Micrometer-scale magnetic imaging of geological samples using a quantum diamond microscope,” Geochem. Geophys., Geosyst. 18, 3254–3267 (2017).
  33. [33]E. V. Levine, M. J. Turner, P. Kehayias, C. A. Hart, N. Langellier, R. Trubko, D. R. Glenn, R. R. Fu, and R. L. Walsworth, “Principles and techniques of the quantum diamond microscope,” Nanophotonics 8, 1945–1973 (2019).arxiv.org/abs/1910.00061
  34. [34]R. Jalilian, L. A. Jauregui, G. Lopez, J. Tian, C. Roecker, M. M. Yazdanpanah, R. W. Cohn, I. Jovanovic, and Y. Chen, “Scanning gate microscopy on graphene: charge inhomogeneity and extrinsic doping,” Nanotechnology 22, 295705 (2011).DOI
  35. [35]F. S. Felt, “Analysis of a microcircuit failure using SQUID and MR current imaging,” International Symposium for Testing and Failure Analysis (ISTFA), 169–177 (2005).DOI
  36. [36]A. Pu, D. Thompson, and G. Bridges, “Location of current carrying failure sites in integrated circuits by magnetic force microscopy at large probe-to-sample separation,” Microelectronic Engineering 86, 16–23 (2009).DOI
  37. [37]J. Henshaw, P. Kehayias, M. S. Ziobari, M. Titze, E. Morissette, K. Watanabe, T. Taniguchi, J. I. A. Li, V. M. Acosta, E. Bielejec, M. P. Lilly, and A. M. Mounce, “Nanoscale solid-state nuclear quadrupole resonance spectroscopy using depth-optimized nitrogen-vacancy ensembles in diamond,” Appl. Phys. Lett. 120, 174002 (2022).DOI
  38. [38]P. Kehayias, E. V. Levine, L. Basso, J. Henshaw, M. Saleh Ziabari, M. Titze, R. Hathli, J. Okoro, D. R. Tibbetts, D. M. Udoni, E. Bielejec, M. P. Lilly, T.-M. Lu, P. D. D. Schwindt, and A. M. Mounce, “Measurement and simulation of the magnetic fields from a 555 timer integrated circuit using a quantum diamond microscope and finite-element analysis,” Phys. Rev. Applied 17, 014021 (2022).DOI
  39. [39]“See supplemental material.”
  40. [40]T. Hingant, J.-P. Tetienne, L. J. Martínez, K. Garcia, D. Ravelosona, J.-F. Roch, and V. Jacques, “Measuring the magnetic moment density in patterned ultrathin ferromagnets with submicrometer resolution,” Phys. Rev. Applied 4, 014003 (2015).
  41. [41]G. J. Abrahams, S. C. Scholten, A. J. Healey, I. O. Robertson, N. Dontschuk, S. Q. Lim, B. C. Johnson, D. A. Simpson, L. C. L. Hollenberg, and J.-P. Tetienne, “An integrated widefield probe for practical diamond nitrogen-vacancy microscopy,” Appl. Phys. Lett. 119, 254002 (2021).DOI
  42. [42]B. J. Roth, N. G. Sepulveda, and J. P. Wikswo, “Using a magnetometer to image a two-dimensional current distribution,” J. Appl. Phys. 65, 361–372 (1989).DOI
  43. [43]K. Chang, A. Eichler, J. Rhensius, L. Lorenzelli, and C. L. Degen, “Nanoscale imaging of current density with a single-spin magnetometer,” Nano Letters 17, 2367–2373 (2017).DOI
  44. [44]D. Broadway, S. Lillie, S. Scholten, D. Rohner, N. Dontschuk, P. Maletinsky, J.-P. Tetienne, and L. Hollenberg, “Improved current density and magnetization reconstruction through vector magnetic field measurements,” Phys. Rev. Applied 14, 024076 (2020).DOI
  45. [45]S. M. Oliver, D. J. Martynovnych, M. J. Turner, D. A. Hopper, R. L. Walsworth, and E. V. Levine, “Vector magnetic current imaging of an 8 nm process node chip and 3d current distributions using the quantum diamond microscope,” arXiv:2202.08135 (2022).DOI
  46. [46]A. Horsley, P. Appel, J. Wolters, J. Achard, A. Tallaire, P. Maletinsky, and P. Treutlein, “Microwave device characterization using a widefield diamond microscope,” Phys. Rev. Applied 10, 044039 (2018).
  47. [47]J. M. Taylor, P. Cappellaro, L. Childress, L. Jiang, D. Budker, P. R. Hemmer, A. Yacoby, R. Walsworth, and M. D. Lukin, “High-sensitivity diamond magnetometer with nanoscale resolution,” Nat. Phys. 4, 810–816 (2008).DOI
  48. [48]M. J. Turner, N. Langellier, R. Bainbridge, D. Walters, S. Meesala, T. M. Babinec, P. Kehayias, A. Yacoby, E. Hu, M. Lončar, R. L. Walsworth, and E. V. Levine, “Magnetic field fingerprinting of integrated-circuit activity with a quantum diamond microscope,” Phys. Rev. Applied 14, 014097 (2020).DOI

Paper details

vlnc:W_3pu32byt
Version
v1 · latest
Published
Fresh — added to the Hub within the last 30 days
Added
License
CC BY 4.0
Usage

views · 8

downloads · 0

A view counts once per browser per day, a download each time the PDF is requested.

  • Physics and Astronomy

Not recorded for this paper

View statement
Looking for related papers…
Contents