Selective Etch for Micromachining Process in Manufacturing Hybrid Microdevices composed of Ni-Mn-Ga and Silicon Layers

The goal of this study is to make selective etch possible for the next generation of MEMS(microelectromechanical systems) devices that are composed Ni-Mn-Ga and silicon layers. Due tothe large magnetic-field-induced strains of Ni-Mn-Ga, sensing and actuating components can be fab-ricated in the Ni-Mn-Ga layers. Other functional components can be manufactured in the silicon layer.Single crystalline Ni-Mn-Ga alloys that are grown by using the Bridgman vertical growth techniquehave so far obtained the largest magnetic field-induced strain (MFIS), a magnetic shape memory(MSM) effect. Similar to silicon wafers, Ni-Mn-Ga wafers are also sliced from crystal-oriented singlecrystalline ingots. To fabricate hybrid MEMS devices such as micromanipulators and robots, lab-on-chip containing micropump manifolds and valves, or vibration energy harvesters, the fabricationprocesses used for MEMS devices will be also used to fabricate components in the Ni-Mn-Ga layer ofthe hybrid MEMS devices. One of the most important processes for MEMS fabrication is the structur-ing of materials by chemical etching. The main goal of this study is to obtain evidence that the etchantetches silicon but not Ni-Mn-Ga and to identify an etchant that etches Ni-Mn-Ga but not silicon. Thepresent paper reports on a novel experiment in dissolving Ni-Mn-Ga alloys. An etchant compositionof 69% HNO3, 98% H2SO4, and CuSO4•5H2O is proposed for dissolving Ni-Mn-Ga alloys and thevariation in the dissolution rate by adjusting the concentrations of HNO3 and ultrapure water (UPW)is demonstrated. This etchant was demonstrated to etch Ni-Mn-Ga but not silicon. The HF+HNO3acidic solution commonly used for etching silicon does not dissolve Ni-Mn-Ga alloys.


Introduction
In recent years, nickel-manganese-gallium (Ni-Mn-Ga) based magnetic shape memory (MSM) alloys haveattractedtheattentionoftheindustryduetotheirveryhighmagnetic-field-inducedstrain(MFIS). The induced strain can be recovered by reorienting external magnetic field applied or by mechanical loading [1 -3]. Unlike thermal shape memory effect, this MSM effect occurs within the martensitic phase, the key to which is the magnetic anisotropy of MSM alloys with tetragonal crystal structures [4 -6]. Regions in the crystals with two different orientations of the short crystallographic c axis that are also along the natural direction of magnetization are called twin boundaries (TBs), illustrated as Fig. 1. During the magnetic-field-induced straining, the TBs move through the crystal until only one variant exists in the material. For the MSM effect to occur, low twining stress and high energy magnetic anisotropy are required [5]. TB movement is inhibited significantly by the grain boundaries in polycrystalline materials; thus, researchers have also applied annealing for grain growth [6 -8], compression [3,9], and matching the grain and sample size [10] to enable high MFIS in polycrystalline Ni-Mn-Ga. Recently, MSM materials have been developed in the forms of single crystals, polycrystals, and nanowires [11 -14] by using Bridgman vertical growth, additive manufacturing [15,16], and electrodeposition. Increasingly more applications in robotics, biomedical, and optics use the Ni-Mn-Ga based MSM alloys such as fast actuators [17,18], micropumps [19], and vibration energy harvesters [20]. Despite of the various developments of Ni-Mn-Ga based MSM alloys aiming at a large MSM effect, single crystalline Ni-Mn-Ga grown by using Bridgman vertical growth still prevails as the most effective technology in terms of the effect of MSM and the potential of mass production. In the semiconductor industry, microelectromechanical systems (MEMS) process technology [21 -24] is applied to the manufacturing of micro-and nanostructured electronic devices, high-precision components, and micro biosensors. This technology allows for the manufacturing of microsized devices and highprecision components with better electrical performance, improved thermal management, and reliable functionality at lower cost and smaller package size. In the past two decades, researchers have worked intensively to advance the MSM materials and their performance. Applying the well-known fabrication processes used in the semiconductor industry, such as selective etching, ion implantation, polishing, and lithography, may offer new methods of MSM fabrication. The proposed process flow of MSM wafer fabrication is illustrated as Fig. 2. Meanwhile, microprocessing of MSM alloys enhance further the working frequency owing to its little hysteresis and high reversibility [26]. The goal is to develop MSM hybrid microdevices that are composed of a Ni-Mn-Ga material layer and a Si layer. These layers are structured using etching and lithographic methods to manufacture actuating, sensing, and functional components in the device. Components fabricated in the Ni-Mn-Ga layers can be as the active regions of actuators, sensors, pumps, valves or grippers [25,27]. Thus it is essential to make selective etch for micromaching process in manufacturing hybrid microdevices composed of both Ni-Mn-Ga and silicon layers. The primary task of this paper is to find such a chemical etchant that dissolves Ni-Mn-Ga but not Si, and to identify a chemical etchant (HF+HNO3) that selectively etches Si but does not dissolve Ni-Mn-Ga. With these motivations in mind, the investigations reported in the present paper aim to achieve the following tasks: 1. To confirm by the experiment that the etchant, an HF+HNO3 acid solution, often used as the chemical solution (selective etch) for silicon, does not dissolve the Ni-Mn-Ga single-crystal sample investigated.
2. An etchant dissolving the Ni-Mn-Ga sample will be identified, and to confirm by the experiment that the etchant for Ni-Mn-Ga dissolution does not dissolve silicon (dissolution rate < 10 mg∕min).
3. The chemical reaction of Ni-Mn-Ga sample with the different compositions of the etchant consisting of H2SO4, HNO3 and CuSO4 will be elucidated.

Experimental Details
A single crystalline Ni-Mn-Ga alloy rod with a diameter of 10 mm and a length of 50 mm was cut into slices with a weight in the range from 0.5 g to 2 g. The slices were added to chemical solutions with different compositions or varied concentrations of the etchant (69% HNO3, 98% H2SO4, and CuSO4•5H2O) at room temperature. The dissolution rate was calculated according to the weight of the sample versus the time of complete dissolution. The sample and experimental parameters are summarized in Tables 1 and 2. To accomplish the abovementioned goals, four experiments were performed as follows:

A. Etching of Si and Ni-Mn-Ga samples with HF+HNO3 acid solution
In Experiment 1, a small piece of the Si Boron-doped p-type (100) sample, with the resistivity of 18-25 Ω .cm, a weight of 1.39 grams and a Ni-Mn-Ga sample with a weight of 1.56 grams was used. An etchant formed by combining 50% HF (10 ml) and 55% HNO3 (10 ml) supplied by Tama Chemicals was used. The Si and Ni-Mn-Ga samples were cleaned by rinsing in ultrapure water (UPW) before etching. UPW had metal traces lower than 1 ng/L inspected routinely on an ICP-MS7500 (Agilent).

C. Rate of Ni-Mn-Ga dissolution with etchants of different compositions.
To develop actuating or sensing components in the Ni-Mn-Ga layer of the hybrid device, the rate of the Ni-Mn-Ga alloy etching must be controlled. Such control may be achieved by changing the ratio of the chemicals in the etchant. Experiment 3 aimed to demonstrate the reactions of the Ni-Mn-Ga sample with different etchants, illustrated as Composition A, B, C and D in Fig. 3. Four compositions were tested at room temperature (23 °C).

D. Fine tuning of the rate of dissolution of Ni-Mn-Ga alloy.
In this experiment, the amount of H2SO4 and CuSO4•5H2O was held constant, whilst the amounts of HNO3 and UPW in the etchant were systematically varied in four steps, referred to as Tests 1 to 4 in Table 2.

Results and Discussion
In Experiment 1, Si reacted to the chemical solution of HNO3 and HF, as shown in Eq 1.
Si + 4HNO3+6HF→ H2SiF6+4NO2↑+4H2O (1) Illustrated as Composition E in Table 1, with a combination of 50% HF (10 ml) and 55% HNO3 (10 ml), the rate of dissolution of the Si sample was found to be 120 mg/min, whereas that of the Ni-Mn-Ga sample was less than 6 mg/min, illustrated as Experiment 1 in Table 1. It concluded that the Ni-Mn-Ga sample did not react effectively with this etchant and remained unetched in the Si etching process commonly used in semiconductor fabrication [27 -29]. The result of Experiment 2 confirmed that the solution made from 69% HNO3, 98% H2SO4, and CuSO4•5H2O dissolved Ni-Mn-Ga but did notetch Si. No obvious reaction with Si was observed during the test. However, this etchant effectively dissolved the Ni-Mn-Ga sample with a rate of 121 mg/min. The following chemical reactions were observed: Nitric acid (HNO3) reacted with nickel, manganese, and gallium to form nickel (II) nitrate, manganese (II) nitrate, and gallium (III) nitrate, respectively, and also to generate hydrogen gas. The reactions are shown as Eq. 2 -4.
Sulfur acid and sulfur-containing species are the good stabilizing agents in this application. During the experiments, the color of the solution changed due to the reaction of Ni (green) and of manganese (yellow) with the sulfuric acid. Gallium is often used as a semiconductor material and is soluble in most of acids and bases. Gallium sulfate, nickel sulfate, and manganese sulfate were formed as shown in Eq. 5 -7. 2Ga + 3H2SO4→Ga2(SO4)3+3H2↑ (5)

84
Manufacturing and Properties of Functional Materials Mn+H2SO4 → MnSO4+H2↑(yellow) Copper (II) sulfate pentahydrate (CuSO4•5H2O), a bright blue crystalline solid, is a hydrate and a metal sulfate that contains copper (II) sulfate. It acts as the stabilizing agent in the reaction with Cu particles, catalyzing the reaction. It also helps to tune the dissolution rate. The detailed reactions are shown in Eq. 8 -10.
CuSO4+H2O → Cu2++(SO4)2−+H2O (8) In Experiment 3, using Composition D, the Ni-Mn-Ga sample was not dissolved effectively, as the dissolution rate was less than 10 mg/min. The dissolution rate ranged from 121 mg/min to 174 mg/min when Compositions A, B, and C were used, as illustrated in Fig. 3. Selecting Composition B as the etchant for Ni-Mn-Ga dissolution, Experiment 4 was conducted to fine tune the composition of the etchant by varying the relative concentrations of the involved chemicals. Four tests, denoted as Tests 1, 2, 3 and 4, were performed on the Ni-Mn-Ga sample. The details of these tests and the results are shown in Table 2 and Fig. 4.  Composition B resulted in a similar dissolution rate of the Ni-Mn-Ga sample, shown as Test 1 in Table 2, that of the Si sample in Experiment 1 (120 mg/min) with a combination of 50% HF (10 ml) and 55% HNO3 (10 ml). In Test 2, the amount of nitric acid (69% HNO3) was increased in volume by 20%. As a result of this modification, the dissolution rate increased to 145 mg/min from 122 mg/min. In Tests 3 and 4, the UPW volume was doubled with respect to the volume used in Tests 1 and 2. When the etchant was diluted, the dissolution rate was reduced to 87mg/min (Test3) from 122mg/min (Test 1), and from 145 mg/min (Test 2) to 102 mg/min (Test 4). Therefore, the rate of dissolution of the Ni-Mn-Ga sample in the etchant (69% HNO3, 98% H2SO4, and CuSO4•5H2O) can be adjusted by modifying the concentrations of the nitric acid or the amount of added UPW. Applying the technology of the semiconductor MEMS micromaching process gives huge flexibility of making complex smart devices that composited of Ni-Mn-Ga alloy as the sensing or actuating components, the active region, thus the inactive regions such as housing, mechanical support, and structures can be fabricated using the standard MEMS micromachining process. Ullakko [30]proposed the approach to develop the MSM devices that consist of the regions of active and inactive, illustrated as in Fig. 5.
In the approach, the sensing or actuating part is developed by using Ni-Mn-Ga alloy, and the housing or other functioning part is made from other materials such as Si substrate. The Ni-Mn-Ga region will be driven by the external magnetic fields.

Conclusions
It was demonstrated that an etchant formed from 69% nitric acid, 98% sulfuric acid, and copper (II) sulfate pentahydrate effectively dissolved the Ni-Mn-Ga alloy sample, but it did not dissolve the Si sample. In contrast, the HF/HNO3 acid solution commonly used for etching Si did not substantially dissolve the Ni-Mn-Ga alloy sample. The results of the experiments presented in this report also illustrated the variations in the rate of Ni-Mn-Ga dissolution with the adjustment of the concentration of nitric acid and amount of added ultrapure water. This result suggests that selective etching, a technology that is also used in semiconductor fabrication, may be possible for Ni-Mn-Ga alloys. These results on selectively etching Si and Ni-Mn-Ga form the basis for developing MSM hybrid microdevices composed of a Ni-Mn-Ga material layer and Si layer, meanwhile, trigger the demand for developing further the micromachining processes for fabricating Ni-Mn-Ga layer such as cleaning, annealing, lithography and bonding.