1. Theory

Definition: Molecular beam epitaxy evaporates elements from effusion cells in ultra-high vacuum so that atoms reach a heated crystalline substrate in straight lines and build the film layer by layer (UC Santa Barbara; Georgia Tech).
Ballistic transport: The mean free path must greatly exceed the source distance; for N $_{2}$ it is about 0.5 km at $1 0^{- 7}$ Torr (Unipress; UC Santa Barbara).
Purity: At $1 0^{- 11}$ Torr a monolayer of residual gas takes about 28 h to arrive, against about 1 s at $1 0^{- 6}$ Torr (Unipress).
Flux: Each Knudsen cell gives a Hertz-Knudsen flux set by its temperature (UC Santa Barbara; Unipress).
Monolayer counting: One RHEED oscillation equals one monolayer in layer-by-layer growth (Unipress).

2. Working Principle

Vacuum system: Cryo, ion and mechanical pumps, an LN $_{2}$ cryoshroud and bakeout reach $1 0^{- 10}$ to $1 0^{- 11}$ Torr (Unipress).
Sources: Effusion cells with individual shutters and drift below 1 per cent per day, valved crackers for As and P, and plasma sources for nitrogen (Unipress).
Flux calibration: Beam equivalent pressure measured by a retractable gauge, about $2$ to $8 \times 1 0^{- 7}$ Torr for Ga (Unipress).
Growth: As-rich GaAs at about 1 ML/s, 0.96 $μ$ m/h, with rate set by the Ga flux (Unipress).

3. Applications

III-V electronics and lasers: HEMT and laser heterostructures with abrupt interfaces (Georgia Tech).
Nitrides and oxides: Plasma-assisted growth, for example GaN at 0.23 $μ$ m/h (Unipress).
Research on growth: In-situ RHEED studies of surface reconstruction and growth mode (Unipress; RHEED page).

Vacuum and Purity

Name	Formula	Related Formulae	Variables/Constants
Ultra-high vacuum	$p_{ba se} = 1 0^{- 10} to 1 0^{- 11} Torr$	$p_{b e am} \approx 1 0^{- 5} Torr$	$U ni p r ess : m ec hani c a l, cr y o an d i o n / T i p u m p s o f 2800, 3000 an d 1200 L/s, bak e n e a r 150^{\circ} C, L N_{2} cr y os h r o u d f r eez in g o u t a t o m s an d r e d u c in g m e m or y e f f ec t s; G eor g ia T ec h : hi g h es t p u r i t y$ $(m y r e a d in g : t h e cr y os h r o u d i s w ha t k ee p s t h e b e am p r ess u r e f r o m r ai s in g t h e ba c k g r o u n d)$
Mean free path	$λ_{N_{2}} \approx 50 cm (1 0^{- 4} Torr), 0.5 km (1 0^{- 7}), 5000 km (1 0^{- 11})$	$λ ≫ L$	$L : so u r ce - s u b s t r a t e d i s t an ce; U ni p r ess : ba l l i s t i c t r an s p or t; U C S an t a B a r ba r a : s am e co n d i t i o n; T U W i e n : λ p ab o u t 5 mm P a$ $(co m p u t e d : T U W i e n^{'} s co n s t an t g i v es 37.5 cm a t 1 0^{- 4} Torr)$
Contamination time	$J = \frac{p}{2 π m k _{B} T}$	$t_{M L} \approx 1 s (1 0^{- 6} Torr), 28 h (1 0^{- 11} Torr)$	$U ni p r ess : 3.2 \times 1 0^{20} c m^{- 2} s^{- 1} Tor r^{- 1} f or m = 40 a t 300 K, ab o u t 1 im p u r i t y p er 1 0^{5} S i a t o m s a t 1 0^{- 11} Torr$ $(co m p u t e d : 3.21 \times 1 0^{20} c m^{- 2} s^{- 1} Tor r^{- 1} r e p r o d u ce d; 1 0^{5} s = 27.8 h)$

Beams, Rate and Uniformity

Name	Formula	Related Formulae	Variables/Constants
Effusion cells and shutters	$J_{ce l l} = f (T_{ce l l}), drift < 1 %/ day$	$crackers: A s_{4} \to A s_{2}; plasma: N_{2}, O_{2}$	$U ni p r ess : K n u d se n ce l l s w i t h in d i v i d u a l s h u tt er s, v a l v e d cr a c k er s f or A s, P, S b, S e, S an d T e; G eor g ia T ec h : ce l l t e m p er a t u r e se t s f l ux, s h u tt er s s w i t c h b e am s; T h er ma l E v a p or a t or p a g e$ $(m y r e a d in g : an M B E ce l l i s a K n u d se n so u r ce in t h e H er t z - K n u d se n se n se)$
Beam equivalent pressure and rate	$B E P_{G a} \approx 2 to 8 \times 1 0^{- 7} Torr (T_{G a} = 830 to 900^{\circ} C)$	$2.67 \overset{˚}{A} /s = 1 ML/s = 0.96 μ m/h$	$U ni p r ess : r e t r a c t ab l e g a ug e, A s - r i c h G a A s w i t h r a t e se t b y G a f l ux, G a N e x am pl e 0.23 μ m/h; G eor g ia T ec h : M B E f or I I I - V e l ec t r o ni cs$ $(co m p u t e d : 2.67 \overset{˚}{A} /s \times 3600 = 0.96 μ m/h)$
Uniformity	$U (\frac{d}{L} = 0.2) \approx 0.93$	$U = (1 + (\frac{d}{L})^{2})^{- 2}$	$d : o f f se t f r o m t h e a x i s; U C S an t a B a r ba r a : U = 1 a t d / L = 0, b e tt er u ni f or mi t y a t l a r g er L b u t l o w er r a t e; T U W i e n : cos^{4} l a w$ $(co m p u t e d : t h e cos^{4} f or m g i v es 0.925 a t d / L = 0.2, ma t c hin g 0.93)$

In-Situ Control

Name	Formula	Related Formulae	Variables/Constants
RHEED oscillations	$v = \frac{1 ML}{τ}$	$E = 5 to 20 keV, θ = 1 to 3^{\circ}$	$τ : osc i l l a t i o n p er i o d; U ni p r ess : o n e p er i o d p er m o n o l a y er, o n l y in 2 D l a y er - b y - l a y er g r o w t h, ab se n t in s t e p f l o w, G a A s (2 \times 4); R H E E D p a g e$ $(co m p u t e d : τ = 1 s a t 1 ML/s)$
Substrate temperature	$T_{s u b} = 200 to 1000^{\circ} C$	$Δ T_{T C} \approx \pm 50^{\circ} C between systems$	$U ni p r ess : c a l ib r a t i o n p o in t s G a A s o x i d e d esor pt i o n ab o u t 560^{\circ} C, S i (111) 7 \times 7 \to 1 \times 1 a t 830^{\circ} C, A l m e l t in g a t 660.3^{\circ} C, t h er m oco u pl es r e a d in g t h e h e a t er$ $(m y r e a d in g : R H E E D t r an s i t i o n s a r e t h e in - s i t u t h er m o m e t er)$
MBE against MOCVD	$MBE: UHV, no gas-phase reactions$	$MOCVD: low-cost optical III-V$	$U C S an t a B a r ba r a : n o t r an s p or t co m pl i c a t i o n s or g a s - p ha se r e a c t i o n s; G eor g ia T ec h : M B E f or e l ec t r o ni cs an d hi g h - e n d l a ser s; M O C V D R e a c t or p a g e$ $(m y r e a d in g : M B E t r a d es t h r o ug h p u t f or in t er f a ce co n t r o l)$

Important Questions

Ques: How does an MBE system grow a crystal?

By aiming atomic beams at a heated substrate in ultra-high vacuum. UC Santa Barbara describes MBE as an evaporation method that brings atoms to the surface one by one, with precise layer-by-layer control of films and doping and no gas-phase reactions. Georgia Tech notes that the long mean free path lets atoms travel in straight lines from the cells to the substrate, that shutters switch each beam and that cell temperature sets the flux. Unipress gives the hardware: effusion cells with individual shutters and drift below 1 per cent per day, valved crackers turning As $_{4}$ into As $_{2}$ , plasma sources for N $_{2}$ and O $_{2}$ , and an LN $_{2}$ cryoshroud. Each cell is a Knudsen source like those on the Thermal Evaporator page (my reading). Reference: UC Santa Barbara Chemical Engineering ChE 142/242, lecture 13 ‘Molecular beam epitaxy’: evaporation bringing atoms to the surface one by one, mean free path much longer than the source distance, layer-by-layer control of films and doping, no transport complications or gas-phase reactions, Knudsen effusion cells with Hertz-Knudsen flux and sticking probability, centreline uniformity U as a function of d/L with U = 1 at d/L = 0 and about 0.93 at d/L = 0.2, uniformity improving with L while the rate falls (fetch-tool summary); W. A. Doolittle, ECE 6450 Lectures 13 and 14 ‘CVD and Epitaxy’, Georgia Institute of Technology: MBE in ultra-high vacuum with the highest purity and finest layer control, dominating III-V electronics and competing in high-end lasers, straight-line travel from cells to the hot substrate, shutters switching the beams and cell temperature setting the flux, MOCVD dominating low-cost optical III-V (fetch-tool summary); Institute of High Pressure Physics PAS (Unipress), crystal growth course 2021-22, lecture 18 ‘Molecular beam epitaxy’: background 1e-10 to 1e-11 torr and about 1e-5 torr inside the beam, N2 mean free path about 50 cm at 1e-4, 0.5 km at 1e-7 and 5000 km at 1e-11 torr, ballistic transport when the mean free path exceeds the source distance, mechanical, cryo and ion/Ti pumps of 2800, 3000 and 1200 L/s, bake at about 150 C, flux J = p/sqrt(2 pi m kB T) with 3.2e20 cm-2 s-1 per torr for m = 40 at 300 K, one monolayer of contamination in about 1 s at 1e-6 torr and 28 h at 1e-11 torr, about 1 impurity per 1e5 Si atoms at 1e-11 torr, Knudsen cells with flux drift below 1 per cent per day and individual shutters, valved crackers turning As4 into As2, plasma sources for N2 and O2, LN2 cryoshroud, Ga BEP about 2e-7 to 8e-7 torr for T_Ga about 830 to 900 C measured by a retractable gauge, As-rich GaAs growth with rate set by Ga flux, 2.67 angstrom/s = 1 ML/s = 0.96 um/h, a GaN example at 0.23 um/h, RHEED at 5 to 20 keV and 1 to 3 degrees with one oscillation period per monolayer, absent in step flow, GaAs (2x4), substrate 200 to 1000 C, calibration points GaAs oxide desorption about 560 C, Si(111) 7x7 to 1x1 at 830 C, Al melting 660.3 C, thermocouples reading the heater with plus or minus 50 C differences between systems (fetch-tool summary); Thermal Evaporator page of this vault.

Ques: Why does MBE need ultra-high vacuum?

For both ballistic beams and purity. Unipress gives background pressures of $1 0^{- 10}$ to $1 0^{- 11}$ $Torr$ , with about $1 0^{- 5}$ $Torr$ inside the beam. Its flux formula $J = p / 2 π m k_{B} T$ gives $3.2 \times 1 0^{20}$ $c m^{- 2} s^{- 1} Tor r^{- 1}$ for mass 40 at 300 K, which I reproduce as $3.21 \times 1 0^{20}$ (computed). Consequently one monolayer of residual gas arrives in about 1 s at $1 0^{- 6}$ $Torr$ but about 28 $h$ at $1 0^{- 11}$ $Torr$ . At $1 0^{- 11}$ $Torr$ impurity incorporation is about 1 atom per $1 0^{5}$ Si atoms. The mean free path for N $_{2}$ grows from about 50 $cm$ at $1 0^{- 4}$ $Torr$ to 5000 $km$ at $1 0^{- 11}$ $Torr$ , far beyond any chamber (Unipress). Reference: Institute of High Pressure Physics PAS (Unipress), crystal growth course 2021-22, lecture 18 ‘Molecular beam epitaxy’: background 1e-10 to 1e-11 torr and about 1e-5 torr inside the beam, N2 mean free path about 50 cm at 1e-4, 0.5 km at 1e-7 and 5000 km at 1e-11 torr, ballistic transport when the mean free path exceeds the source distance, mechanical, cryo and ion/Ti pumps of 2800, 3000 and 1200 L/s, bake at about 150 C, flux J = p/sqrt(2 pi m kB T) with 3.2e20 cm-2 s-1 per torr for m = 40 at 300 K, one monolayer of contamination in about 1 s at 1e-6 torr and 28 h at 1e-11 torr, about 1 impurity per 1e5 Si atoms at 1e-11 torr, Knudsen cells with flux drift below 1 per cent per day and individual shutters, valved crackers turning As4 into As2, plasma sources for N2 and O2, LN2 cryoshroud, Ga BEP about 2e-7 to 8e-7 torr for T_Ga about 830 to 900 C measured by a retractable gauge, As-rich GaAs growth with rate set by Ga flux, 2.67 angstrom/s = 1 ML/s = 0.96 um/h, a GaN example at 0.23 um/h, RHEED at 5 to 20 keV and 1 to 3 degrees with one oscillation period per monolayer, absent in step flow, GaAs (2x4), substrate 200 to 1000 C, calibration points GaAs oxide desorption about 560 C, Si(111) 7x7 to 1x1 at 830 C, Al melting 660.3 C, thermocouples reading the heater with plus or minus 50 C differences between systems (fetch-tool summary).

Ques: How is the MBE growth rate set and measured?

By the group III cell temperature, checked by beam pressure and RHEED. Unipress grows GaAs As-rich, so the rate is set by the Ga flux with no Ga desorption. It measures Ga beam equivalent pressure with a retractable gauge, about $2$ to $8 \times 1 0^{- 7}$ $Torr$ for cell temperatures of 830 to 900 $^{\circ} C$ . Its typical rate is 2.67 $\overset{˚}{A} /s$ , or 1 $ML/s$ , which is 0.96 $μ m/h$ (computed). RHEED gives the rate directly: one oscillation period equals one monolayer, so $v = 1 ML / τ$ . This works only in two-dimensional layer-by-layer growth, not step flow (Unipress). A 1 s period therefore means 1 $ML/s$ (computed). The method is described on the RHEED page. Reference: Institute of High Pressure Physics PAS (Unipress), crystal growth course 2021-22, lecture 18 ‘Molecular beam epitaxy’: background 1e-10 to 1e-11 torr and about 1e-5 torr inside the beam, N2 mean free path about 50 cm at 1e-4, 0.5 km at 1e-7 and 5000 km at 1e-11 torr, ballistic transport when the mean free path exceeds the source distance, mechanical, cryo and ion/Ti pumps of 2800, 3000 and 1200 L/s, bake at about 150 C, flux J = p/sqrt(2 pi m kB T) with 3.2e20 cm-2 s-1 per torr for m = 40 at 300 K, one monolayer of contamination in about 1 s at 1e-6 torr and 28 h at 1e-11 torr, about 1 impurity per 1e5 Si atoms at 1e-11 torr, Knudsen cells with flux drift below 1 per cent per day and individual shutters, valved crackers turning As4 into As2, plasma sources for N2 and O2, LN2 cryoshroud, Ga BEP about 2e-7 to 8e-7 torr for T_Ga about 830 to 900 C measured by a retractable gauge, As-rich GaAs growth with rate set by Ga flux, 2.67 angstrom/s = 1 ML/s = 0.96 um/h, a GaN example at 0.23 um/h, RHEED at 5 to 20 keV and 1 to 3 degrees with one oscillation period per monolayer, absent in step flow, GaAs (2x4), substrate 200 to 1000 C, calibration points GaAs oxide desorption about 560 C, Si(111) 7x7 to 1x1 at 830 C, Al melting 660.3 C, thermocouples reading the heater with plus or minus 50 C differences between systems (fetch-tool summary); Reflection High-Energy Electron Diffraction (RHEED) page of this vault.

Ques: What is the mean free path at 10^-4 Torr, and which value is more likely?

About 40 to 50 $cm$ ; the two sources differ by about 30 per cent. Unipress gives about 50 $cm$ for N $_{2}$ at 300 K and $1 0^{- 4}$ $Torr$ . TU Wien’s rule $λ p \approx 5$ mm Pa gives 37.5 $cm$ at the same pressure (computed). TU Wien states its constant only approximately and without naming the gas. The Unipress value names N $_{2}$ and 300 K, so it is more likely correct for that case (my reading). The difference does not matter for MBE: at the operating background of $1 0^{- 10}$ to $1 0^{- 11}$ $Torr$ , either rule gives a mean free path of kilometres, far beyond the source distance. UC Santa Barbara states the same condition, $λ ≫ L$ . Reference: Institute of High Pressure Physics PAS (Unipress), crystal growth course 2021-22, lecture 18 ‘Molecular beam epitaxy’: background 1e-10 to 1e-11 torr and about 1e-5 torr inside the beam, N2 mean free path about 50 cm at 1e-4, 0.5 km at 1e-7 and 5000 km at 1e-11 torr, ballistic transport when the mean free path exceeds the source distance, mechanical, cryo and ion/Ti pumps of 2800, 3000 and 1200 L/s, bake at about 150 C, flux J = p/sqrt(2 pi m kB T) with 3.2e20 cm-2 s-1 per torr for m = 40 at 300 K, one monolayer of contamination in about 1 s at 1e-6 torr and 28 h at 1e-11 torr, about 1 impurity per 1e5 Si atoms at 1e-11 torr, Knudsen cells with flux drift below 1 per cent per day and individual shutters, valved crackers turning As4 into As2, plasma sources for N2 and O2, LN2 cryoshroud, Ga BEP about 2e-7 to 8e-7 torr for T_Ga about 830 to 900 C measured by a retractable gauge, As-rich GaAs growth with rate set by Ga flux, 2.67 angstrom/s = 1 ML/s = 0.96 um/h, a GaN example at 0.23 um/h, RHEED at 5 to 20 keV and 1 to 3 degrees with one oscillation period per monolayer, absent in step flow, GaAs (2x4), substrate 200 to 1000 C, calibration points GaAs oxide desorption about 560 C, Si(111) 7x7 to 1x1 at 830 C, Al melting 660.3 C, thermocouples reading the heater with plus or minus 50 C differences between systems (fetch-tool summary); TU Wien Institute of Solid State Physics, ‘Physik und Technologie Duenner Schichten / Thin Film Technology’, chapter 2: mean free path lambda p about 5 mm Pa, Knudsen cell emission and cos^4 flat-substrate law (fetch-tool summary); UC Santa Barbara Chemical Engineering ChE 142/242, lecture 13 ‘Molecular beam epitaxy’: evaporation bringing atoms to the surface one by one, mean free path much longer than the source distance, layer-by-layer control of films and doping, no transport complications or gas-phase reactions, Knudsen effusion cells with Hertz-Knudsen flux and sticking probability, centreline uniformity U as a function of d/L with U = 1 at d/L = 0 and about 0.93 at d/L = 0.2, uniformity improving with L while the rate falls (fetch-tool summary).

Ques: How uniform is an MBE film, and how is substrate temperature known?

Uniformity follows the evaporation geometry, and temperature needs in-situ calibration. UC Santa Barbara gives centreline uniformity as a function of $d / L$ , with $U = 1$ on axis and about 0.93 at $d / L = 0.2$ ; uniformity improves with distance $L$ but the rate falls. The flat-substrate $cos^{4}$ law of TU Wien gives 0.925 at $d / L = 0.2$ (computed), in agreement. For temperature, Unipress warns that thermocouples read the heater, not the substrate, and that differences of about plus or minus 50 $^{\circ} C$ between systems are standard. It calibrates against GaAs oxide desorption near 560 $^{\circ} C$ , the Si(111) $7 \times 7 \to 1 \times 1$ transition at 830 $^{\circ} C$ and Al melting at 660.3 $^{\circ} C$ . RHEED transitions thus serve as the thermometer (my reading). Reference: UC Santa Barbara Chemical Engineering ChE 142/242, lecture 13 ‘Molecular beam epitaxy’: evaporation bringing atoms to the surface one by one, mean free path much longer than the source distance, layer-by-layer control of films and doping, no transport complications or gas-phase reactions, Knudsen effusion cells with Hertz-Knudsen flux and sticking probability, centreline uniformity U as a function of d/L with U = 1 at d/L = 0 and about 0.93 at d/L = 0.2, uniformity improving with L while the rate falls (fetch-tool summary); TU Wien Institute of Solid State Physics, ‘Physik und Technologie Duenner Schichten / Thin Film Technology’, chapter 2: mean free path lambda p about 5 mm Pa, Knudsen cell emission and cos^4 flat-substrate law (fetch-tool summary); Institute of High Pressure Physics PAS (Unipress), crystal growth course 2021-22, lecture 18 ‘Molecular beam epitaxy’: background 1e-10 to 1e-11 torr and about 1e-5 torr inside the beam, N2 mean free path about 50 cm at 1e-4, 0.5 km at 1e-7 and 5000 km at 1e-11 torr, ballistic transport when the mean free path exceeds the source distance, mechanical, cryo and ion/Ti pumps of 2800, 3000 and 1200 L/s, bake at about 150 C, flux J = p/sqrt(2 pi m kB T) with 3.2e20 cm-2 s-1 per torr for m = 40 at 300 K, one monolayer of contamination in about 1 s at 1e-6 torr and 28 h at 1e-11 torr, about 1 impurity per 1e5 Si atoms at 1e-11 torr, Knudsen cells with flux drift below 1 per cent per day and individual shutters, valved crackers turning As4 into As2, plasma sources for N2 and O2, LN2 cryoshroud, Ga BEP about 2e-7 to 8e-7 torr for T_Ga about 830 to 900 C measured by a retractable gauge, As-rich GaAs growth with rate set by Ga flux, 2.67 angstrom/s = 1 ML/s = 0.96 um/h, a GaN example at 0.23 um/h, RHEED at 5 to 20 keV and 1 to 3 degrees with one oscillation period per monolayer, absent in step flow, GaAs (2x4), substrate 200 to 1000 C, calibration points GaAs oxide desorption about 560 C, Si(111) 7x7 to 1x1 at 830 C, Al melting 660.3 C, thermocouples reading the heater with plus or minus 50 C differences between systems (fetch-tool summary).

Ques: When is MBE chosen over MOCVD?

When interface control and purity outweigh throughput. Georgia Tech states that MBE gives the highest purity and finest layer control, dominates the III-V electronics market and competes in high-end lasers, while MOCVD dominates low-cost optical III-V technology (MOCVD Reactor page). UC Santa Barbara lists MBE’s advantages as layer-by-layer control of films and doping without transport complications or gas-phase reactions. Unipress quotes about 1 $ML/s$ , or 0.96 $μ m/h$ , for GaAs and 0.23 $μ m/h$ for a GaN example. In-situ RHEED shows each monolayer as it forms. In my reading, the shutters let compositions switch within a monolayer, which is why abrupt heterojunctions for HEMTs favour MBE. Reference: W. A. Doolittle, ECE 6450 Lectures 13 and 14 ‘CVD and Epitaxy’, Georgia Institute of Technology: MBE in ultra-high vacuum with the highest purity and finest layer control, dominating III-V electronics and competing in high-end lasers, straight-line travel from cells to the hot substrate, shutters switching the beams and cell temperature setting the flux, MOCVD dominating low-cost optical III-V (fetch-tool summary); UC Santa Barbara Chemical Engineering ChE 142/242, lecture 13 ‘Molecular beam epitaxy’: evaporation bringing atoms to the surface one by one, mean free path much longer than the source distance, layer-by-layer control of films and doping, no transport complications or gas-phase reactions, Knudsen effusion cells with Hertz-Knudsen flux and sticking probability, centreline uniformity U as a function of d/L with U = 1 at d/L = 0 and about 0.93 at d/L = 0.2, uniformity improving with L while the rate falls (fetch-tool summary); Institute of High Pressure Physics PAS (Unipress), crystal growth course 2021-22, lecture 18 ‘Molecular beam epitaxy’: background 1e-10 to 1e-11 torr and about 1e-5 torr inside the beam, N2 mean free path about 50 cm at 1e-4, 0.5 km at 1e-7 and 5000 km at 1e-11 torr, ballistic transport when the mean free path exceeds the source distance, mechanical, cryo and ion/Ti pumps of 2800, 3000 and 1200 L/s, bake at about 150 C, flux J = p/sqrt(2 pi m kB T) with 3.2e20 cm-2 s-1 per torr for m = 40 at 300 K, one monolayer of contamination in about 1 s at 1e-6 torr and 28 h at 1e-11 torr, about 1 impurity per 1e5 Si atoms at 1e-11 torr, Knudsen cells with flux drift below 1 per cent per day and individual shutters, valved crackers turning As4 into As2, plasma sources for N2 and O2, LN2 cryoshroud, Ga BEP about 2e-7 to 8e-7 torr for T_Ga about 830 to 900 C measured by a retractable gauge, As-rich GaAs growth with rate set by Ga flux, 2.67 angstrom/s = 1 ML/s = 0.96 um/h, a GaN example at 0.23 um/h, RHEED at 5 to 20 keV and 1 to 3 degrees with one oscillation period per monolayer, absent in step flow, GaAs (2x4), substrate 200 to 1000 C, calibration points GaAs oxide desorption about 560 C, Si(111) 7x7 to 1x1 at 830 C, Al melting 660.3 C, thermocouples reading the heater with plus or minus 50 C differences between systems (fetch-tool summary); MOCVD Reactor page of this vault.

References

Institute of High Pressure Physics PAS (Unipress), “Molecular beam epitaxy,” crystal growth course 2021-22, lecture 18. Available: https://korfu.unipress.waw.pl/~stach/cg-2021-22/cgm_w18.pdf
UC Santa Barbara, ChE 142/242, “Lecture 13: molecular beam epitaxy.” Available: https://sites.chemengr.ucsb.edu/~ceweb/courses/che142242/pdfs/lecture_13_chex42.pdf
W. A. Doolittle, “CVD and epitaxy,” ECE 6450 Lectures 13 and 14, Georgia Institute of Technology. Available: https://alan.ece.gatech.edu/ECE6450/Lectures/ECE6450L13and14-CVD%20and%20Epitaxy.pdf
TU Wien Institute of Solid State Physics, “Physik und Technologie duenner Schichten (Thin film technology),” chapter 2. Available: https://static.ifp.tuwien.ac.at/homepages/Personen/duenne_schichten/pdf/t_p_tf_chapter2.pdf
Reflection High-Energy Electron Diffraction (RHEED) page of this vault.
MOCVD Reactor page of this vault.
Thermal Evaporator page of this vault.

Quartz 5

Explorer

Molecular Beam Epitaxy (MBE) System