Wet Etch Bench

1. Theory

• Definition: A wet etch bench removes films by immersion in liquid etchants, through transport of reactants, surface reaction and removal of products (Darling).
• Figures of merit: Selectivity $S=r_2/r_1$ and anisotropy $A=1-r_{lat}/r_{vert}$; most wet etches are isotropic and undercut the mask (KTH; Nayak et al.).
• Oxide etching: HF dissolves SiO${}_2$ as H${}_2$SiF${}_6$; NH${}_4$F buffering keeps the fluoride supply steady (Darling; KTH).
• Crystal anisotropy: Alkaline etchants attack (100) far faster than (111), giving 54.74 degree walls (Darling; Nayak et al.).
• Etch stops: Heavy boron doping slows alkaline etching 10 to 50 times (Darling).

2. Working Principle

• Baths: BOE at controlled temperature, KOH at 80 ${}^{\circ }$C with agitation, TMAH at 90 ${}^{\circ }$C, phosphoric acid at 160 ${}^{\circ }$C (General Chemical; Darling; Berkeley).
• Temperature control: BOE changes by up to 100 angstrom/min per degree, so baths are held to $\pm 0.5$ ${}^{\circ }$C (General Chemical).
• Masks: Resist for BOE, oxide for hot phosphoric acid, nitride for KOH (Williams et al.; Darling).
• After etching: Rinse and dry as on the Spin Rinse Dryer and IPA Vapor Dryer page (my reading).

3. Applications

• Oxide patterning and HF dips: BOE and dilute HF (General Chemical; RCA Wet Bench page).
• MEMS bulk micromachining: KOH and TMAH cavities and membranes with boron stops (Darling).
• Nitride strip: Hot phosphoric acid after LOCOS (KTH; Berkeley).

Figures of Merit and Oxide Etching

Name	Formula	Related Formulae	Variables/Constants
Figures of merit	$S=\frac{r_2}{r_1}$	$A=1-\frac{r_{lat}}{r_{vert}}$	$\begin{array}{r}KTH:A=0isotropic,A=1anisotropic;\\ Darling:isotropicetchequalinall\\ directions;Nayaketal.:wetetching\\ generallyisotropicwithundercut\end{array}$ $\begin{array}{r}(myreading:awetisotropicetch\\ undercutsthemaskbyaboutitsdepth)\end{array}$
Oxide in HF	$Si{O}_2+6HF\to H_{2}Si{F}_6+2H_{2}O$	$r_{BOE}(21{}^{\circ }C)=1020\text{to}1120(5:1),860\text{to}930(6:1),510\text{to}580(10:1)\mathring{\mathrm{A}}/min$	$\begin{array}{r}DarlingandKTH:reaction;General\\ Chemical:20:1at265to315and50:1at\\ 125to145\mathring{\mathrm{A}}/min,upto100\mathring{\mathrm{A}}/minper\\ degree,controltoplusorminus0.5{}^{\circ }C;\\ KTH:N{H}_{4}Fbufferprevents{F}^{-}depletion\end{array}$ $\begin{array}{r}(computed:10:1over5:1is545/1070=\\ 0.51,matchingWilliamsetal{.}^{\prime }sexactly\\ half)\end{array}$
Surfactant BOE	$\gamma =19\text{to}23\text{against}85\text{to}90dyn/cm$	$\text{over-etch}:20\text{to}30\%\to \approx 10\%$	$\begin{array}{r}GeneralChemical:betterwettingand\\ uniformity,nopredip,BOEstopsatthe\\ Si/Si{O}_{2}interface\end{array}$ $\begin{array}{r}(myreading:betterwettingofsmall\\ openingsiswhatcutstheneeded\\ over-etch)\end{array}$

Anisotropic Silicon Etching

Name	Formula	Related Formulae	Variables/Constants
Anisotropic silicon etch	$r_{(111)}:r_{(110)}:r_{(100)}\approx 1:600:400\text{(KOH)}$	$r_{(100)}\approx 1\mu m/min\text{(KOH, 80}{}^{\circ }C)$	$\begin{array}{r}Darling:TMAH1:10to1:35,EDP1:35,\\ aminegallate1:50to1:100,(111)stop\\ planes;Berkeley:TMAHabout50\mu m/hat90\\ {}^{\circ }C\end{array}$ $\begin{array}{r}(computed:1\mu m/minis60\mu m/h,against\\ Berkele{y}^{\prime }s50\mu m/hforTMAH)\end{array}$
V-groove geometry	$\theta =54.74^{\circ }$	$d=D-\frac{2h}{\tan 54.7^{\circ }}$	$d:floorwidth$ $D:maskopening$ $h:depth;Nayaketal.;Darling$ $\begin{array}{r}(computed:a100\mu \mathrm{m}deepetchnarrowsthe\\ floorby141.8\mu \mathrm{m})\end{array}$
Boron etch stop	$N_B>10^{20}\mathrm{c}{\mathrm{m}}^{-3}\Rightarrow r_{KOH}÷20$	$N_B>7\times 10^{19}\mathrm{c}{\mathrm{m}}^{-3}\Rightarrow r_{EDP}÷50$	$\begin{array}{r}Darling:NaOH10timesabove3\times 10^{20},TMAH\\ 10timesabove10^{20},HNAfasterwith\\ doping;Williamsetal.:KOHusedtoa\\ heavilydopedp-typestop\end{array}$ $\begin{array}{r}(myreading:thestopmakesmembrane\\ thicknesssetbythedopingdepth)\end{array}$

Isotropic Silicon, Nitride and Metals

Name	Formula	Related Formulae	Variables/Constants
Isotropic silicon (HNA)	$Si+HN{O}_3+6HF\to H_{2}Si{F}_6+HN{O}_2+H_{2}O+H_2$	$HF:HN{O}_3:H_{2}O=3:2:5$	$\begin{array}{r}Darling:oxidationbyHN{O}_{3}thendissolution\\ byHF,30to50angstromresidualoxidein\\ HN{O}_3-richmixtures\end{array}$ $\begin{array}{r}(computed:Si,F,NandObalance;H\\ balancesas7oneachside)\end{array}$
Nitride in hot phosphoric acid	$r_{SiN}=0.0872e^{0.0386T}\mathring{\mathrm{A}}/min$	$r_{SiN}(160{}^{\circ }C)\approx 42\mathring{\mathrm{A}}/min$	$\begin{array}{r}Tin{}^{\circ }C;Berkeley:85percent{H}_{3}P{O}_4;\\ Williamsetal.:verytemperature\\ sensitive,resistattackedsooxidemasks\\ used;KTH:oneofthefewwetetchesfor\\ nitride\end{array}$ $(computed:3.9percentperdegree)$
Masks and metals	$r_{Al}(50{}^{\circ }C)=6000\mathring{\mathrm{A}}/min$	$r_{KOH}:Si{O}_2\approx 20\mathring{\mathrm{A}}/min,S{i}_3{N}_4\approx 14\text{angstrom/h}$	$\begin{array}{r}Berkeley:TransenetypeA;Darling:EDP\\ maskedbyoxide,nitrideandnoblemetals\\ butetchingAl,TMAHsparingSi{O}_{2}andAl\end{array}$ $\begin{array}{r}(computed:KOHselectivityof(100)Si\\ overoxideabout500)\end{array}$

Important Questions

Ques: How fast does BOE etch thermal oxide, and how sensitive is it?

About 50 to 110 $nm/min$ depending on dilution. General Chemical lists rates at 21 ${}^{\circ }\mathrm{C}$ of 1020 to 1120 $\mathring{\mathrm{A}}/min$ for 5:1, 860 to 930 for 6:1, 510 to 580 for 10:1, 265 to 315 for 20:1 and 125 to 145 for 50:1. The 10:1 to 5:1 ratio is 0.51 (computed), matching Williams et al.’s statement that 10:1 BHF etches at exactly half the 5:1 rate. A 1 ${}^{\circ }\mathrm{C}$ change can shift the rate by as much as 100 $\mathring{\mathrm{A}}/min$, so baths are held to plus or minus 0.5 ${}^{\circ }\mathrm{C}$. KTH explains that the NH${}_4$F buffer prevents fluoride depletion and reduces resist attack. Surfactant BOE cuts the needed over-etch from 20 to 30 per cent to about 10 per cent (General Chemical). Short HF dips appear on the RCA Wet Bench page, and the liquid-free alternative on the Vapor HF and Plasma Dry Clean Tool page. Reference: General Chemical Corporation, ‘Buffered oxide etchants (BOE)’ brochure: 40 per cent NH4F with 49 per cent HF, thermal oxide rates at 21 C of 1020 to 1120 (5:1), 860 to 930 (6:1), 735 to 805 (7:1), 510 to 580 (10:1), 265 to 315 (20:1) and 125 to 145 (50:1) angstrom/min, a 1 C change altering rate by as much as 100 angstrom/min, control to plus or minus 0.5 C, surfactant BOE at 19 to 23 dyn/cm against 85 to 90 cutting over-etch from 20 to 30 per cent to about 10 per cent, BOE stopping at the Si/SiO2 interface (fetch-tool summary); K. R. Williams, K. Gupta and M. Wasilik, ‘Etch rates for micromachining processing - Part II’, Journal of Microelectromechanical Systems 12(6), 761-778 (2003), text only: 5:1 BHF from 40 per cent NH4F and 49 per cent HF, 10:1 BHF etching thermal oxide at exactly half the 5:1 rate, HF oxide rate linear with concentration from 10:1 to 100:1, PSG etching much faster than LTO, hot phosphoric acid very temperature sensitive and attacking resist so oxide masks are used, KOH etching oxides and LPCVD nitride slowly (fetch-tool summary); KTH Royal Institute of Technology, Lecture 7 ‘Etching’ (course slides): selectivity S = r2/r1, anisotropy A = 1 - r_lat/r_vert with A = 0 isotropic and 1 anisotropic, NH4F buffer preventing F- depletion and lowering resist attack, SiO2 + 6HF -> H2SiF6 + 2H2O, HNA isotropic, hot phosphoric acid for Si3N4, KOH anisotropic for MEMS, wet etching mostly isotropic with poor resolution and harmful chemicals (fetch-tool summary); RCA Wet Bench page of this vault; Vapor HF and Plasma Dry Clean Tool page of this vault.

Ques: Why does KOH etch silicon anisotropically?

Because the etch rate depends strongly on crystal plane. Darling orders the rates (100) > (110) > (111), gives KOH about 1 $\mu m/min$ on (100) and a (111):(110):(100) ratio of about 1:600:400, so the slow (111) planes act as stop planes at 54.74 degrees. Nayak et al. give the floor width of a pit as $d=D-2h/\tan 54.7^{\circ }$; a 100 $\mu \mathrm{m}$ deep etch narrows it by 141.8 $\mu \mathrm{m}$ (computed). Darling lists TMAH at 1:10 to 1:35, EDP at 1:35 and amine gallate at 1:50 to 1:100. KOH etches oxide at about 20 $\mathring{\mathrm{A}}/min$ and nitride at about 14 angstrom/h, so nitride is the better mask. The oxide selectivity is about 500 (computed). Reference: R. B. Darling, ‘Wet Etching’, EE-527 MicroFabrication lecture slides, University of Washington: transport of reactants, surface reaction and transport of products, SiO2 + 6HF -> H2SiF6 + 2H2O, HNA Si + HNO3 + 6HF -> H2SiF6 + HNO2 + H2O + H2 with anodic, cathodic, oxidation and oxide-dissolution steps, 3:2:5 HF:HNO3:H2O example, 30 to 50 angstrom residual oxide in HNO3-rich mixtures, rates (100) > (110) > (111) with (111) as stop planes and 54.74 degrees, KOH about 1 um/min on (100) with (111):(110):(100) about 1:600:400, TMAH about 1 um/min at 90 C with (111):(100) 1:10 to 1:35, EDP 14 to 36 um/h from 70 to 97 C with 1:35, amine gallate 1.7 um/min at 118 C, boron etch stops KOH 20x above 1e20 cm-3, NaOH 10x above 3e20, EDP 50x above 7e19, TMAH 10x above 1e20, HNA faster with heavier doping, EDP masked by SiO2, Si3N4, Au, Cr, Ag, Cu, Ta but etching Al, TMAH not etching SiO2 or Al significantly, KOH etching Si3N4 at about 14 angstrom/h and oxide at about 20 angstrom/min, KOH at 80 C with agitation, isotropic etch equal in all directions (fetch-tool summary); A. P. Nayak, Logeeswaran VJ and M. S. Islam (UC Davis), ‘Wet and dry etching’: (100) Si pits with 54.7 degree walls, floor width d = D - 2h/tan(54.7), wet etching generally isotropic with undercut and usually faster than dry etching (fetch-tool summary).

Ques: How does a boron etch stop work, and what concentrations are needed?

Heavy p-type doping slows alkaline etchants sharply. Darling gives KOH slowing 20 times above $10^{20}$ $\mathrm{c}{\mathrm{m}}^{-3}$, NaOH 10 times above $3\times 10^{20}$, EDP 50 times above $7\times 10^{19}$ and TMAH 10 times above $10^{20}$ $\mathrm{c}{\mathrm{m}}^{-3}$, while HNA instead speeds up with heavier doping. Williams et al. describe KOH as used for etching n-type silicon down to a heavily doped p-type stop. In my reading, this lets membrane thickness be set by the boron diffusion depth rather than by timing, which is far more reproducible. EDP’s stop is the strongest of those listed, but EDP etches aluminium, whereas TMAH spares SiO${}_2$ and Al (Darling). Reference: R. B. Darling, ‘Wet Etching’, EE-527 MicroFabrication lecture slides, University of Washington: transport of reactants, surface reaction and transport of products, SiO2 + 6HF -> H2SiF6 + 2H2O, HNA Si + HNO3 + 6HF -> H2SiF6 + HNO2 + H2O + H2 with anodic, cathodic, oxidation and oxide-dissolution steps, 3:2:5 HF:HNO3:H2O example, 30 to 50 angstrom residual oxide in HNO3-rich mixtures, rates (100) > (110) > (111) with (111) as stop planes and 54.74 degrees, KOH about 1 um/min on (100) with (111):(110):(100) about 1:600:400, TMAH about 1 um/min at 90 C with (111):(100) 1:10 to 1:35, EDP 14 to 36 um/h from 70 to 97 C with 1:35, amine gallate 1.7 um/min at 118 C, boron etch stops KOH 20x above 1e20 cm-3, NaOH 10x above 3e20, EDP 50x above 7e19, TMAH 10x above 1e20, HNA faster with heavier doping, EDP masked by SiO2, Si3N4, Au, Cr, Ag, Cu, Ta but etching Al, TMAH not etching SiO2 or Al significantly, KOH etching Si3N4 at about 14 angstrom/h and oxide at about 20 angstrom/min, KOH at 80 C with agitation, isotropic etch equal in all directions (fetch-tool summary); K. R. Williams, K. Gupta and M. Wasilik, ‘Etch rates for micromachining processing - Part II’, Journal of Microelectromechanical Systems 12(6), 761-778 (2003), text only: 5:1 BHF from 40 per cent NH4F and 49 per cent HF, 10:1 BHF etching thermal oxide at exactly half the 5:1 rate, HF oxide rate linear with concentration from 10:1 to 100:1, PSG etching much faster than LTO, hot phosphoric acid very temperature sensitive and attacking resist so oxide masks are used, KOH etching oxides and LPCVD nitride slowly (fetch-tool summary).

Ques: How is silicon etched isotropically in wet chemistry?

With HNA, an oxidise-then-dissolve mixture. Darling gives the overall reaction $Si+HN{O}_3+6HF\to H_{2}Si{F}_6+HN{O}_2+H_{2}O+H_2$, which balances (computed). It splits into steps: holes oxidise silicon, $S{i}^{2+}$ forms Si(OH)${}_2$ and then SiO${}_2$, and HF dissolves the oxide as H${}_2$SiF${}_6$. An example mixture is HF:HNO${}_3$:H${}_2$O of 3:2:5 by volume. HNO${}_3$-rich mixtures leave 30 to 50 angstrom of residual oxide, which in my reading means the oxide dissolution is then rate-limiting. KTH classes HNA as isotropic, and Darling notes that isotropic etching proceeds at the same rate in all directions. HNA etches faster in heavily doped silicon, the opposite of KOH, so it cannot use a boron etch stop and is timed instead (Darling; my reading). Reference: R. B. Darling, ‘Wet Etching’, EE-527 MicroFabrication lecture slides, University of Washington: transport of reactants, surface reaction and transport of products, SiO2 + 6HF -> H2SiF6 + 2H2O, HNA Si + HNO3 + 6HF -> H2SiF6 + HNO2 + H2O + H2 with anodic, cathodic, oxidation and oxide-dissolution steps, 3:2:5 HF:HNO3:H2O example, 30 to 50 angstrom residual oxide in HNO3-rich mixtures, rates (100) > (110) > (111) with (111) as stop planes and 54.74 degrees, KOH about 1 um/min on (100) with (111):(110):(100) about 1:600:400, TMAH about 1 um/min at 90 C with (111):(100) 1:10 to 1:35, EDP 14 to 36 um/h from 70 to 97 C with 1:35, amine gallate 1.7 um/min at 118 C, boron etch stops KOH 20x above 1e20 cm-3, NaOH 10x above 3e20, EDP 50x above 7e19, TMAH 10x above 1e20, HNA faster with heavier doping, EDP masked by SiO2, Si3N4, Au, Cr, Ag, Cu, Ta but etching Al, TMAH not etching SiO2 or Al significantly, KOH etching Si3N4 at about 14 angstrom/h and oxide at about 20 angstrom/min, KOH at 80 C with agitation, isotropic etch equal in all directions (fetch-tool summary); KTH Royal Institute of Technology, Lecture 7 ‘Etching’ (course slides): selectivity S = r2/r1, anisotropy A = 1 - r_lat/r_vert with A = 0 isotropic and 1 anisotropic, NH4F buffer preventing F- depletion and lowering resist attack, SiO2 + 6HF -> H2SiF6 + 2H2O, HNA isotropic, hot phosphoric acid for Si3N4, KOH anisotropic for MEMS, wet etching mostly isotropic with poor resolution and harmful chemicals (fetch-tool summary).

Ques: How is silicon nitride wet-etched?

In hot phosphoric acid, one of the few wet etches for it. Berkeley’s lab manual gives 85 per cent H${}_3$PO${}_4$ at 160 ${}^{\circ }\mathrm{C}$ with a rate of $0.0872e^{0.0386T}$ $\mathring{\mathrm{A}}/min$, T in ${}^{\circ }\mathrm{C}$. At 160 ${}^{\circ }\mathrm{C}$ that is about 42 $\mathring{\mathrm{A}}/min$, rising 3.9 per cent per degree (computed). Williams et al. confirm that the rate is very temperature sensitive, that their bath was probably a few degrees hot, and that phosphoric acid attacks photoresist, so an oxide mask is used. KTH lists it as the nitride etchant. At about 4 $nm/min$, a 100 $\mathrm{nm}$ nitride takes roughly 25 $\min$ (computed). A degree of temperature error therefore changes the strip time by about 4 per cent, so bath control matters as much as for BOE (my reading). Reference: Marvell NanoLab (UC Berkeley) lab manual chapter 1.9 ‘VLSI etchants’, based on Williams and Muller (JMEMS 1996) and Williams, Gupta and Wasilik (JMEMS 2003): phosphoric acid (85 per cent, 160 C) rate 0.0872 exp(0.0386 T) angstrom/min with T in C, KOH 1000 g in 2 L at 80 C, TMAH 3:2 25 per cent at 90 C about 50 um/h, Transene Al etchant type A (80 per cent H3PO4, 5 per cent HNO3, 5 per cent CH3COOH) 6000 angstrom/min at 50 C, poly etchant about 100 angstrom/s, rates accurate to plus or minus 10 per cent plus or minus 5 angstrom/min, wet etches at about 20 C (fetch-tool summary); K. R. Williams, K. Gupta and M. Wasilik, ‘Etch rates for micromachining processing - Part II’, Journal of Microelectromechanical Systems 12(6), 761-778 (2003), text only: 5:1 BHF from 40 per cent NH4F and 49 per cent HF, 10:1 BHF etching thermal oxide at exactly half the 5:1 rate, HF oxide rate linear with concentration from 10:1 to 100:1, PSG etching much faster than LTO, hot phosphoric acid very temperature sensitive and attacking resist so oxide masks are used, KOH etching oxides and LPCVD nitride slowly (fetch-tool summary); KTH Royal Institute of Technology, Lecture 7 ‘Etching’ (course slides): selectivity S = r2/r1, anisotropy A = 1 - r_lat/r_vert with A = 0 isotropic and 1 anisotropic, NH4F buffer preventing F- depletion and lowering resist attack, SiO2 + 6HF -> H2SiF6 + 2H2O, HNA isotropic, hot phosphoric acid for Si3N4, KOH anisotropic for MEMS, wet etching mostly isotropic with poor resolution and harmful chemicals (fetch-tool summary).

Ques: What is the TMAH etch rate at 90 C, and which figure is more likely?

About 50 to 60 $\mu m/h$; the two sources nearly agree. Darling gives TMAH at about 1 $\mu m/min$ at 90 ${}^{\circ }\mathrm{C}$, which is 60 $\mu m/h$ (computed). The Berkeley lab manual, based on Williams and co-workers’ measurements, gives about 50 $\mu m/h$ for 3:2 diluted 25 per cent TMAH at 90 ${}^{\circ }\mathrm{C}$. The Berkeley value is more likely for that recipe, because it names the concentration and comes from measured data. Darling’s figure is a rounded general value (my reading). Berkeley states an accuracy of plus or minus 10 per cent, so the gap is about twice that uncertainty. Rate also depends on concentration and temperature, as Darling’s EDP series from 14 to 36 $\mu m/h$ over 70 to 97 ${}^{\circ }\mathrm{C}$ shows. Reference: R. B. Darling, ‘Wet Etching’, EE-527 MicroFabrication lecture slides, University of Washington: transport of reactants, surface reaction and transport of products, SiO2 + 6HF -> H2SiF6 + 2H2O, HNA Si + HNO3 + 6HF -> H2SiF6 + HNO2 + H2O + H2 with anodic, cathodic, oxidation and oxide-dissolution steps, 3:2:5 HF:HNO3:H2O example, 30 to 50 angstrom residual oxide in HNO3-rich mixtures, rates (100) > (110) > (111) with (111) as stop planes and 54.74 degrees, KOH about 1 um/min on (100) with (111):(110):(100) about 1:600:400, TMAH about 1 um/min at 90 C with (111):(100) 1:10 to 1:35, EDP 14 to 36 um/h from 70 to 97 C with 1:35, amine gallate 1.7 um/min at 118 C, boron etch stops KOH 20x above 1e20 cm-3, NaOH 10x above 3e20, EDP 50x above 7e19, TMAH 10x above 1e20, HNA faster with heavier doping, EDP masked by SiO2, Si3N4, Au, Cr, Ag, Cu, Ta but etching Al, TMAH not etching SiO2 or Al significantly, KOH etching Si3N4 at about 14 angstrom/h and oxide at about 20 angstrom/min, KOH at 80 C with agitation, isotropic etch equal in all directions (fetch-tool summary); Marvell NanoLab (UC Berkeley) lab manual chapter 1.9 ‘VLSI etchants’, based on Williams and Muller (JMEMS 1996) and Williams, Gupta and Wasilik (JMEMS 2003): phosphoric acid (85 per cent, 160 C) rate 0.0872 exp(0.0386 T) angstrom/min with T in C, KOH 1000 g in 2 L at 80 C, TMAH 3:2 25 per cent at 90 C about 50 um/h, Transene Al etchant type A (80 per cent H3PO4, 5 per cent HNO3, 5 per cent CH3COOH) 6000 angstrom/min at 50 C, poly etchant about 100 angstrom/s, rates accurate to plus or minus 10 per cent plus or minus 5 angstrom/min, wet etches at about 20 C (fetch-tool summary).

References

1. R. B. Darling, “Wet etching,” EE-527 MicroFabrication lecture notes, University of Washington. Available: https://mail.wecanfigurethisout.org/NANO/lecture_notes/Microfabrication_Supporting_materials_files/R_B_Darling - Washington U - WetEtching.pdf
2. KTH Royal Institute of Technology, “Lecture 7: etching,” course slides. Available: https://www.kth.se/social/upload/510f795cf276544e1ddda13f/Lecture%207%20Etching.pdf
3. General Chemical Corporation, “Buffered oxide etchants,” product brochure. Available: https://litho.nano.cnr.it/wp-content/datasheets/BufferedOxideEchants_brochure.pdf
4. Marvell NanoLab, UC Berkeley, “Chapter 1.9: VLSI etchants,” lab manual. Available: https://nanolab.berkeley.edu/labmanual/chap1/1.09vslietch.pdf
5. K. R. Williams, K. Gupta and M. Wasilik, “Etch rates for micromachining processing - Part II,” Journal of Microelectromechanical Systems, vol. 12, no. 6, pp. 761-778, 2003. Available: https://www.seas.upenn.edu/~nanosop/documents/Etchratesformicromachiningprocessing.pdf
6. A. P. Nayak, Logeeswaran VJ and M. S. Islam, “Wet and dry etching,” UC Davis. Available: https://www.ece.ucdavis.edu/~anayakpr/Papers/Wet and Dry Etching_submitted.pdf
7. RCA Wet Bench page of this vault.
8. Vapor HF and Plasma Dry Clean Tool page of this vault.

Related pages in this vault: RCA Wet Bench, Vapor HF and Plasma Dry Clean Tool, Spin Rinse Dryer and IPA Vapor Dryer, Stylus Profiler, Spectroscopic Ellipsometer, Plasma Asher (Barrel and Downstream), Parallel-Plate Reactive Ion Etcher (RIE), Deep Reactive Ion Etcher (Bosch DRIE).