📘 Lecture 7 — Geometric meaning of partial derivative
📖 Overview: This lecture explores the geometric interpretation of partial derivatives for functions of two variables. It explains how partial derivatives represent slopes of tangent lines to curves formed by intersecting a surface with planes of constant x or y. The lecture also covers higher-order partial derivatives, mixed partial derivatives, Laplace’s equation, and Euler’s theorem on the equality of mixed partials under continuity conditions.
🗂️ Topics Covered
Geometric meaning of partial derivatives, partial derivatives and their definitions using limits, interpretation of ∂z/∂x as slope of tangent to curve of intersection with y=constant, interpretation of ∂z/∂y as slope of tangent to curve of intersection with x=constant, second-order partial derivatives and their notations, mixed partial derivatives f_xy and f_yx, worked examples computing partial derivatives, Laplace’s equation ∇²f = 0, Euler’s theorem on equality of mixed partials, and the advantage of using Euler’s theorem to simplify computation order.
📝 Lecture Summary
Geometric meaning of partial derivative
Let z = f(x, y) be a function of two variables. The graph of f is a surface in three-dimensional space. Consider a point P on this graph with coordinates (x₀, y₀, f(x₀, y₀)). If a point starts at P and moves along the surface such that y remains constant, the path traced is the curve of intersection between the surface z = f(x, y) and the plane y = constant. On this curve, ∂z/∂x represents the derivative of f with respect to x while y is held constant.
🔑 Definition — ∂z/∂x: The slope of the tangent to the curve of intersection of the surface z = f(x, y) and the plane y = constant at point P.
Similarly, ∂z/∂y is the gradient of the tangent at point P to the curve formed by the intersection of the surface z = f(x, y) and the plane x = constant. These two tangent lines together define the tangent plane to the surface at point P.
Partial Derivatives
Let z = f(x, y) be a function of two variables defined on a domain D. For a given change ∆x in x, keeping y fixed, the change ∆z in z is given by:
∆z = f(x + ∆x, y) – f(x, y)
If the ratio ∆z/∆x = [f(x + ∆x, y) – f(x, y)]/∆x approaches a finite limit as ∆x → 0, then this limit is called the partial derivative of f with respect to x.
Similarly, for a change ∆y in y, keeping x fixed: ∆z = f(x, y + ∆y) – f(x, y)
If the ratio [f(x, y + ∆y) – f(x, y)]/∆y approaches a finite limit as ∆y → 0, this limit is called the partial derivative of f with respect to y.
Partial Derivatives of Higher Orders
The first-order partial derivatives f_x and f_y of a function f of two variables x and y are themselves functions of x and y, and may possess derivatives. These are called second-order partial derivatives.
The four second-order partial derivatives are:
- ∂/∂x(∂f/∂x) = ∂²f/∂x² = ∂/∂x(f_x) = (f_x)_x = f_xx = f_x²
- ∂/∂y(∂f/∂x) = ∂²f/∂y∂x = ∂/∂y(f_x) = (f_x)_y = f_xy
- ∂/∂x(∂f/∂y) = ∂²f/∂x∂y = ∂/∂x(f_y) = (f_y)_x = f_yx
- ∂/∂y(∂f/∂y) = ∂²f/∂y² = ∂/∂y(f_y) = (f_y)_y = f_yy = f_y²
Thus, there are four second-order partial derivatives for a function z = f(x, y). The partial derivatives f_xy and f_yx are called mixed second partials and are not equal in general. Partial derivatives of order more than two can be defined similarly.
🔑 Definition — Mixed partial derivatives: Derivatives where differentiation is performed with respect to different variables, such as f_xy (first differentiate with respect to x, then y) and f_yx (first differentiate with respect to y, then x).
Example
Given z = arc sin(x/y), compute partial derivatives.
📐 Formula: ∂z/∂x = 1/√(y² - x²)
📐 Formula: ∂z/∂y = -x/(y√(y² - x²))
📌 Example: For the mixed partial ∂²z/∂y∂x: ∂²z/∂y∂x = ∂/∂y(∂z/∂x) = -1/2(y² - x²)^(-3/2)·2y = -y/(y² - x²)^(3/2)
For ∂²z/∂x∂y: ∂²z/∂x∂y = ∂/∂x(∂z/∂y) = -1 · [√(y² - x²) - x·(1/2)(y² - x²)^(-1/2)·(-2x)]/(y² - x²) = -y/(y² - x²)^(3/2)
Hence, ∂²z/∂x∂y = ∂²z/∂y∂x in this case.
Example
Given f(x, y) = x cos y + y eˣ
📌 Example: Compute partial derivatives: ∂f/∂x = cos y + y eˣ ∂f/∂y = -x sin y + eˣ
∂²f/∂y∂x = ∂/∂y(∂f/∂x) = -sin y + eˣ ∂²f/∂x² = ∂/∂x(∂f/∂x) = y eˣ ∂²f/∂x∂y = ∂/∂x(∂f/∂y) = -sin y + eˣ ∂²f/∂y² = ∂/∂y(∂f/∂y) = -x cos y
Laplace’s Equation
For a function w = f(x, y, z), the equation:
📐 Formula: ∂²f/∂x² + ∂²f/∂y² + ∂²f/∂z² = 0
is called Laplace’s equation. Functions satisfying this equation are called harmonic functions.
Example
Given f(x, y) = eˣ sin y + eʸ cos x
📌 Example: Compute second derivatives: ∂f/∂x = eˣ sin y - eʸ sin x ∂²f/∂x² = eˣ sin y - eʸ cos x
∂f/∂y = eˣ cos y + eʸ cos x ∂²f/∂y² = -eˣ sin y + eʸ cos x
Adding: ∂²f/∂x² + ∂²f/∂y² = (eˣ sin y - eʸ cos x) + (-eˣ sin y + eʸ cos x) = 0
Thus, this function satisfies Laplace’s equation.
💡 Why this matters: Laplace’s equation appears in physics (electrostatics, fluid dynamics, heat conduction) — functions satisfying it describe steady-state phenomena.
Euler’s Theorem
The mixed derivative theorem (Euler’s theorem):
If f(x, y) and its partial derivatives f_x, f_y, f_xy, and f_yx are defined throughout an open region containing a point (a, b) and are all continuous at (a, b), then:
📐 Formula: f_xy(a, b) = f_yx(a, b)
🔑 Definition — Euler’s theorem: Under conditions of continuity, the order of differentiation does not matter for mixed partial derivatives.
Advantage of Euler’s theorem
Given w = xy + eʸ/(y² + 1), the symbol ∂²w/∂x∂y tells us to differentiate first with respect to y and then with respect to x. However, by Euler’s theorem, we can postpone differentiation with respect to y and differentiate first with respect to x for a quicker solution.
📌 Example: ∂w/∂x = y, then ∂²w/∂y∂x = 1
This is simpler than differentiating first with respect to y and then with respect to x.
⭐ Key Takeaways
The partial derivative ∂z/∂x geometrically represents the slope of the tangent line to the curve formed by intersecting the surface with a plane of constant y. Similarly, ∂z/∂y represents the slope of the tangent to the curve formed by intersecting with a plane of constant x. For a function of two variables, there are four second-order partial derivatives: f_xx, f_xy, f_yx, and f_yy. Euler’s theorem states that if all second-order partial derivatives are continuous in an open region, then the mixed partials are equal: f_xy = f_yx. This allows flexibility in the order of differentiation, often simplifying calculations. Laplace’s equation ∂²f/∂x² + ∂²f/∂y² + ∂²f/∂z² = 0 is a fundamental partial differential equation with applications in physics and engineering.
🧠 Quick Revision Questions
- What is the geometric meaning of ∂z/∂x for a function z = f(x, y)?
- How many second-order partial derivatives exist for a function of two variables? List them with their notations.
- State Euler’s theorem (the mixed derivative theorem) and the conditions under which it holds.
- Verify whether f(x, y) = eˣ sin y + eʸ cos x satisfies Laplace’s equation.
- For f(x, y) = x cos y + y eˣ, compute f_xy and f_yx and verify if they are equal.
📘 Lecture 8 — More About Euler Theorem Chain Rule
📖 Overview: This lecture explores the concept of changing the order of differentiation for higher-order partial derivatives, demonstrating that the order does not affect the result when the function and its partial derivatives are continuous. It then transitions to the chain rule for functions of one and several variables, showing how to differentiate composite functions systematically.
🗂️ Topics Covered
The lecture covers the invariance of the order of differentiation for higher-order partial derivatives under continuity conditions, with examples illustrating how strategic ordering can simplify complex calculations. It then introduces the chain rule for functions of one variable (w=f(x), x=g(t)) and extends it to functions of two variables (w=f(x,y), x=g(t), y=h(t)). Multiple examples demonstrate both substitution and chain rule methods for computing derivatives.
📝 Lecture Summary
Order of differentiation
In general, the order of differentiation in an nth order partial derivative can be changed without affecting the final result whenever the function and all its partial derivatives of order ≤ n are continuous. For example, if f and its partial order derivatives of the first, second, and third orders are continuous on an open set, then at each point of the set: f_{xyy} = f_{yxy} = f_{yyx}. In another notation: ∂³f/(∂y²∂x) = ∂³f/(∂y∂x∂y) = ∂³f/(∂x∂y²).
For a function f(x,y) = y²x⁴eˣ + 2, if we need ∂⁵f/(∂y³∂x²), differentiating first w.r.t. x then w.r.t. y involves many steps. But differentiating with respect to y first and then x second makes the calculation easier. ∂⁵f/(∂x²∂y³) = 0.
🔑 Definition — Order of differentiation: The sequence in which partial derivatives are taken can be rearranged without changing the result when all relevant partial derivatives are continuous.
📐 Formula: For continuous partial derivatives: f_{xyy} = f_{yxy} = f_{yyx}
📌 Example: f(x,y) = (x+y)/(x−y) fₓ(x,y) = [(x−y)(1) − (x+y)(1)]/(x−y)² = −2y/(x−y)² fᵧ(x,y) = [(x−y)(1) − (x+y)(−1)]/(x−y)² = 2x/(x−y)²
📌 Example: f(x,y) = x³e^(−y) + y³sec(xy) fₓ(x,y) = 3x²e^(−y) + y³sec(xy)tan(xy) fᵧ(x,y) = −x³e^(−y) + 3y²sec(xy)
📌 Example: f(x,y) = x²ye^(xy) fₓ(x,y) = 2xye^(xy) + x²y²e^(xy) = xye^(xy)(2+xy) fₓ(1,1) = (1)(1)e¹[2+1] = 3e
fᵧ(x,y) = x²e^(xy) + x³ye^(xy) = x²e^(xy)(1+xy) fᵧ(1,1) = (1)(1)e¹[1+1] = 2e
📌 Example: f(x,y) = x²cos(xy) fₓ(x,y) = 2xcos(xy) − x²ysin(xy) fₓ(½, π) = 2(½)cos(π/2) − (½)²(π)sin(π/2) = 0 − π/4 = −π/4
fᵧ(x,y) = −x³sin(xy) fᵧ(½, π) = −(½)³sin(π/2) = −1/8
📌 Example: w = (4x − 3y + 2z)⁵ ∂w/∂x = 20(4x−3y+2z)⁴ ∂²w/(∂y∂x) = −240(4x−3y+2z)³ ∂³w/(∂z∂y∂x) = −1440(4x−3y+2z)² ∂⁴w/(∂z²∂y∂x) = −576(4x−3y+2z)
Chain Rule in function of One variable
Given that w = f(x) and x = g(t), we find dw/dt as follows: From w = f(x), we get dw/dx. From x = g(t), we get dx/dt. Then dw/dt = (dw/dx)(dx/dt).
📌 Example: w = x + 4, x = sin t By Substitution: w = sin t + 4, dw/dt = cos t By Chain Rule: dw/dx = 1, dx/dt = cos t, dw/dt = 1·cos t = cos t
Chain rule in function of one variable (extended): If y is a function of u, u is a function of v, v is a function of w, w is a function of z, z is a function of x, then ultimately y is function of x, so by chain rule: dy/dx = (dy/du)(du/dv)(dv/dw)(dw/dz)(dz/dx)
w = f(x,y), x = g(t), y = f(t)
For a function w = f(x,y) where x and y are functions of t, the chain rule gives: dw/dt = (∂w/∂x)(dx/dt) + (∂w/∂y)(dy/dt)
📐 Formula: Chain rule for two variables: dw/dt = (∂w/∂x)(dx/dt) + (∂w/∂y)(dy/dt)
📌 Example BY SUBSTITUTION: w = xy, x = cos t, y = sin t w = cos t sin t = ½ sin 2t dw/dt = ½(cos 2t)(2) = cos 2t
📌 Example BY CHAIN RULE: w = xy, x = cos t, y = sin t ∂w/∂x = y = sin t, ∂w/∂y = x = cos t dx/dt = −sin t, dy/dt = cos t dw/dt = (sin t)(−sin t) + (cos t)(cos t) = −sin²t + cos²t = cos 2t
📌 Example: z = 3x²y³, x = t⁴, y = t² ∂z/∂x = 6xy³, ∂z/∂y = 9x²y² dx/dt = 4t³, dy/dt = 2t dz/dt = (6xy³)(4t³) + (9x²y²)(2t) = 6(t⁴)(t⁶)(4t³) + 9(t⁸)(t⁴)(2t) = 24t¹³ + 18t¹³ = 42t¹³
📌 Example: z = √(1 + x − 2xy⁴), x = ln t, y = t ∂z/∂x = 1/(2√(1+x−2xy⁴)) ∂z/∂y = 1/(2√(1+x−2xy⁴)) · (−8xy³) = −4xy³/√(1+x−2xy⁴) dx/dt = 1/t, dy/dt = 1 dz/dt = (∂z/∂x)(dx/dt) + (∂z/∂y)(dy/dt) = 1/(2√(1+x−2xy⁴)) + −4xy³/√(1+x−2xy⁴) = 1/(2t)√(1+x−2xy⁴) − 4xy³/√(1+x−2xy⁴) = (1/√(1+x−2xy⁴))[1/(2t) − 4xy³] = (1/√(1+ln t − 2t⁴ln t))[1/(2t) − 4(ln t)t³]
📌 Example: z = ln(2x² + y), x = √t, y = t^(2/3) ∂z/∂x = (1/(2x²+y))·4x = 4x/(2x²+y) ∂z/∂y = 1/(2x²+y) dx/dt = 1/(2√t), dy/dt = (2/3)t^(−1/3)
⭐ Key Takeaways
The order of differentiation for higher-order partial derivatives can be freely interchanged when the function and all its partial derivatives up to that order are continuous, which can significantly simplify complex derivative calculations. The chain rule for functions of one variable, dw/dt = (dw/dx)(dx/dt), extends naturally to functions of multiple variables where each intermediate variable depends on a single independent variable t. For w = f(x,y) with x = g(t) and y = h(t), the chain rule is dw/dt = (∂w/∂x)(dx/dt) + (∂w/∂y)(dy/dt), which can be verified by direct substitution. Strategic choice of differentiation order and the chain rule are powerful techniques for efficiently computing higher-order and composite derivatives.
🧠 Quick Revision Questions
- Under what condition can the order of partial differentiation be changed without affecting the result?
- For f(x,y) = x²ye^(xy), what is fₓ(1,1)?
- What is dw/dt if w = f(x) and x = g(t)?
- State the chain rule formula for w = f(x,y) where x = g(t) and y = h(t).
- For w = xy with x = cos t and y = sin t, verify that dw/dt = cos 2t using both substitution and chain rule methods.
📘 Lecture 9 — Examples
📖 Overview: This lecture focuses on applying the chain rule for functions of multiple variables through worked examples. It reviews the chain rule for functions with one intermediate variable, extends it to functions with multiple intermediate variables and multiple independent variables, and demonstrates how to compute mixed partial derivatives using this technique.
🗂️ Topics Covered
The lecture begins by revising an example from Lecture 8 of the chain rule for w = f(x,y,z) with x, y, z as functions of t. It then covers the chain rule for w = f(x) where x = g(r, s), followed by the case of w = f(x,y) where x = g(r, s) and y = h(r, s), and finally w = f(x,y,z) where x = g(r, s), y = h(r, s), z = k(r, s). The lecture concludes with a general statement of the chain rule for functions of many variables and a complex example involving partial derivatives of w = ln(e^r + e^s + e^t + e^u).
📝 Lecture Summary
Revision of Previous Example
The lecture first revises the chain rule for w = f(x,y,z) where x = g(t), y = f(t), z = h(t). The total derivative with respect to t is:
dw/dt = (∂w/∂x)(dx/dt) + (∂w/∂y)(dy/dt) + (∂w/∂z)(dz/dt)
📌 Example: w = x² + y + z + 4, x = eᵗ, y = cos t, z = t + 4
- ∂w/∂x = 2x, ∂w/∂y = 1, ∂w/∂z = 1
- dx/dt = eᵗ, dy/dt = -sin t, dz/dt = 1
- dw/dt = (2x)(eᵗ) + (1)(-sin t) + (1)(1) = 2(eᵗ)(eᵗ) - sin t + 1 = 2e²ᵗ - sin t + 1
w = f(x) with x = g(r, s)
For w = f(x) where x = g(r, s), x is an intermediate variable. The partial derivatives are:
∂w/∂r = (dw/dx)(∂x/∂r)
∂w/∂s = (dw/dx)(∂x/∂s)
📌 Example: w = sin x + x², x = 3r + 4s
- dw/dx = cos x + 2x
- ∂x/∂r = 3, ∂x/∂s = 4
- ∂w/∂r = (cos x + 2x)(3) = 3 cos(3r+4s) + 18r + 24s
- ∂w/∂s = (cos x + 2x)(4) = 4 cos(3r+4s) + 24r + 32s
w = f(x,y) with x = g(r,s), y = h(r,s)
For the dependent variable w = f(x,y) where x and y are intermediate variables and r and s are independent variables:
∂w/∂r = (∂w/∂x)(∂x/∂r) + (∂w/∂y)(∂y/∂r)
∂w/∂s = (∂w/∂x)(∂x/∂s) + (∂w/∂y)(∂y/∂s)
w = f(x,y,z) with x = g(r,s), y = h(r,s), z = k(r,s)
Extending to three intermediate variables: ∂w/∂r = (∂w/∂x)(∂x/∂r) + (∂w/∂y)(∂y/∂r) + (∂w/∂z)(∂z/∂r) ∂w/∂s = (∂w/∂x)(∂x/∂s) + (∂w/∂y)(∂y/∂s) + (∂w/∂z)(∂z/∂s)
📌 Example: w = x + 2y + z², x = r/s, y = r² + ln s, z = 2r
- ∂w/∂x = 1, ∂w/∂y = 2, ∂w/∂z = 2z
- ∂x/∂r = 1/s, ∂y/∂r = 2r, ∂z/∂r = 2
- ∂w/∂r = (1)(1/s) + (2)(2r) + (2z)(2) = 1/s + 4r + 4(2r) = 1/s + 12r
- ∂x/∂s = -r/s², ∂y/∂s = 1/s, ∂z/∂s = 0
- ∂w/∂s = (1)(-r/s²) + (2)(1/s) + (2z)(0) = 2/s - r/s²
Remembering the Different Forms of the Chain Rule
The best approach is to draw a tree diagram placing the dependent variable on top, intermediate variables in the middle, and the selected independent variable at the bottom. To find the derivative, start at the dependent variable and read down each branch to the independent variable, calculating and multiplying the derivatives along the branch, then add the products from different branches.
General Chain Rule for Functions of Many Variables:
Suppose ω = f(x, y, ..., υ) is a differentiable function of variables x, y, ..., υ (a finite set), and x, y, ..., υ are differentiable functions of p, q, ..., t (another finite set). Then ω is a differentiable function of the variables p through t, and the partial derivatives of ω with respect to these variables are given by:
∂ω/∂p = (∂ω/∂x)(∂x/∂p) + (∂ω/∂y)(∂y/∂p) + ... + (∂ω/∂υ)(∂υ/∂p)
💡 Why this matters: This general form can be remembered as the dot product of two vectors: (∂ω/∂x, ∂ω/∂y, ..., ∂ω/∂υ) and (∂x/∂p, ∂y/∂p, ..., ∂υ/∂p).
Advanced Example: Partial Derivatives of w = ln(e^r + e^s + e^t + e^u)
Given w = ln(e^r + e^s + e^t + e^u), we can rewrite as e^w = e^r + e^s + e^t + e^u.
Taking partial derivatives:
- e^w·∂w/∂r = e^r → ∂w/∂r = e^(r-w)
- e^w·∂w/∂s = e^s → ∂w/∂s = e^(s-w)
- e^w·∂w/∂t = e^t → ∂w/∂t = e^(t-w)
- e^w·∂w/∂u = e^u → ∂w/∂u = e^(u-w)
Now computing mixed partial derivatives:
- ∂²w/∂r∂s: w_rs = e^(r-w)(-w_s) = -e^(r-w)·e^(s-w) = -e^(r+s-2w)
- ∂³w/∂r∂s∂t: w_rst = -e^(r+s-2w)(-2w_t) = 2e^(r+s-2w)·e^(t-w) = 2e^(r+s+t-3w)
- ∂⁴w/∂r∂s∂t∂u: w_rstu = 2e^(r+s+t-3w)(-3w_u) = -6e^(r+s+t-3w)·e^(u-w) = -6e^(r+s+t+u-4w)
⭐ Key Takeaways
The chain rule for functions of many variables requires systematically drawing tree diagrams and tracing all paths from the dependent variable to the independent variable. When a dependent variable w depends on intermediate variables x, y, z, which themselves depend on independent variables r, s, the partial derivative ∂w/∂r equals the sum of (∂w/∂x)(∂x/∂r) + (∂w/∂y)(∂y/∂r) + (∂w/∂z)(∂z/∂r). For mixed partial derivatives, each successive differentiation requires careful application of the chain rule, substituting previously computed partial derivatives. The general form of the chain rule can be remembered as the dot product of the gradient vector of w with respect to intermediate variables and the vector of partial derivatives of intermediate variables with respect to the chosen independent variable.
🧠 Quick Revision Questions
- For w = f(x,y) where x = g(r,s) and y = h(r,s), write the formula for ∂w/∂r.
- In the example w = x² + y + z + 4 with x = eᵗ, y = cos t, z = t + 4, what is the value of dw/dt at t = 0?
- For w = x + 2y + z² with x = r/s, y = r² + ln s, z = 2r, calculate ∂w/∂r and ∂w/∂s.
- What is the general formula for the chain rule when ω = f(x, y, ..., υ) and x, y, ..., υ are functions of p?
- If w = ln(e^r + e^s + e^t + e^u), find ∂w/∂r and ∂²w/∂r∂s.
📘 Lecture 10 — Introduction to Vectors
📖 Overview: This lecture introduces the fundamental concept of vectors as quantities possessing both magnitude and direction, distinguishing them from scalars. It covers vector representation, basic operations (addition, scalar multiplication, dot and cross products), and their geometric interpretations. Understanding vectors is essential for describing physical quantities like force and velocity, and they form the foundation for many topics in physics and engineering.
🗂️ Topics Covered
The lecture begins by defining vectors and their representation through directed line segments. It then covers vector properties such as magnitude, unit vectors, and equality. The core operations are explained: vector addition (resultant), scalar multiplication, the scalar (dot) product with its component form and angle calculation, and the vector (cross) product with its determinant formula and geometric interpretation of area. Key special cases for parallel and perpendicular vectors are also examined.
📝 Lecture Summary
Introduction to vectors
Some physical measurements are determined solely by their magnitude, but quantities like force and velocity require both magnitude and direction for a complete description. Quantities with direction and magnitude are usually represented by arrows pointing in the direction of the action, with the length of the arrow representing the magnitude relative to a chosen unit. A vector in the plane is a directed line segment. Vectors are usually denoted by a single boldface letter or a letter with an arrow, e.g., the vector from point A to point B is written as AB.
Magnitude and Unit Vector
The magnitude (or length) of a vector v is the length of the directed line segment representing it. A unit vector is any vector whose magnitude is 1. The unit vector in the direction of v is denoted by v̂ and is given by v̂ = v / |v|.
Addition of Vectors
Vectors are added geometrically by connecting them head-to-tail. The resultant vector is the single vector drawn from the tail of the first vector to the head of the last vector. The resultant r = a + b + c + ...
Equal, Opposite, and Parallel Vectors
- Equal vectors have the same magnitude and direction.
- Opposite vectors have the same magnitude but opposite direction.
- Parallel vectors are scalar multiples of each other: b = λa, where λ is a non-zero scalar.
Vector Representation in Rectangular Components
Any vector can be expressed in terms of its components along the x, y, and z axes using the unit vectors i, j, and k: r = x i + y j + z k.
Addition and Subtraction in Component Form
If a = a₁i + a₂j + a₃k and b = b₁i + b₂j + b₃k, then:
- a + b = (a₁ + b₁)i + (a₂ + b₂)j + (a₃ + b₃)k
- a - b = (a₁ - b₁)i + (a₂ - b₂)j + (a₃ - b₃)k
Key idea: corresponding components (i, j, k) are added or subtracted.
Multiplication of a Vector by a Scalar
Multiplying a vector a by a scalar λ results in a parallel vector λa. The direction remains the same if λ > 0, and is reversed if λ < 0.
Scalar (Dot) Product
The scalar product (dot product) of vectors a and b is the number: a·b = |a| |b| cos θ, where θ is the angle between a and b. It is commutative (a·b = b·a).
🔑 Definition — Commutative Law (for dot product): a·b = b·a. This means the order of the vectors in a dot product does not change the result.
Results of Scalar Product:
- a ⟂ b (perpendicular): If a is perpendicular to b, then a·b = 0. Similarly, i·j = j·k = k·i = 0.
- a ∥ b (parallel): If a is parallel to b, then a·b = |a||b|. The dot product of a vector with itself is a·a = |a|², so i·i = j·j = k·k = 1.
Dot Product in Component Form
If a = a₁i + a₂j + a₃k and b = b₁i + b₂j + b₃k, then their dot product is calculated by multiplying corresponding components and summing: a·b = a₁b₁ + a₂b₂ + a₃b₃.
💡 Why this matters: This component form allows for the algebraic calculation of the scalar product without needing to know the angle between the vectors.
Angle Between Two Vectors
The angle θ between two non-zero vectors a and b is given by: θ = cos⁻¹( a·b / (|a||b|) ). The value of θ lies in the interval [0, π].
🔑 Formula: θ = cos⁻¹( a·b / (|a||b|) ) → The angle is the inverse cosine of the dot product divided by the product of the magnitudes.
📌 Example: Find the angle between a = i - 2j - 2k and b = 6i + 3j + 2k. Step 1: Calculate a·b = (1)(6) + (-2)(3) + (-2)(2) = 6 - 6 - 4 = -4. Step 2: Calculate |a| = √(1² + (-2)² + (-2)²) = √9 = 3. Step 3: Calculate |b| = √(6² + 3² + 2²) = √49 = 7. Step 4: θ = cos⁻¹(-4 / (3*7)) = cos⁻¹(-4/21) ≈ 1.76 rad.
Perpendicular (Orthogonal) Vectors
Vectors a and b are perpendicular if and only if a·b = 0. This is an "if and only if" relationship, meaning it works both ways.
Vector Projection
The vector projection of b onto a is given by: proj_a b = ( (b·a) / (a·a) ) a. The scalar component of b in the direction of a is |b| cos θ = b·(a/|a|).
📌 Example: Find the vector projection of b = 6i + 3j + 2k onto a = 1i - 2j - 2k. b·a = -4 (from previous example). a·a = 1² + (-2)² + (-2)² = 9. proj_a b = (-4/9)(1i - 2j - 2k) = (-4/9)i + (8/9)j + (8/9)k. The scalar component of b in the direction of a is -4/3.
Cross Product of Two Vectors
The vector product (cross product) of two non-zero vectors a and b is: a × b = (|a||b| sin θ) n, where n is a unit vector perpendicular to both a and b, determined by the right-hand rule.
🔑 Definition — Right-hand rule: To find the direction of a × b, point your fingers from a towards b through the angle θ, and your right thumb points in the direction of n (and thus a × b).
Results of Cross Product:
- If a ∥ b, then a × b = 0. So a × a = 0 and i × i = j × j = k × k = 0.
- If a ⟂ b, then a × b = |a||b| n.
- i × j = k, j × i = -k ; j × k = i, k × j = -i ; k × i = j, i × k = -j. Note: the cross product is not commutative.
Cross Product as Area
The magnitude of the cross product |a × b| is equal to the area of the parallelogram determined by a and b, where |a| is the base and |b||sinθ| is the height.
Cross Product in Component Form
The cross product can be computed using a determinant.
📐 Formula: a × b = det| i j k; a₁ a₂ a₃; b₁ b₂ b₃ | → This determinant expansion gives the components of the cross product.
📌 Example: Find the cross product a × b for a = 2i + 1j + 1k and b = -4i + 3j + 1k. Step 1: Set up the determinant: | i j k | | 2 1 1 | | -4 3 1 | Step 2: Calculate: i(11 - 13) - j(21 - 1(-4)) + k(23 - 1(-4)) = i(1 - 3) - j(2 + 4) + k(6 + 4) = -2i - 6j + 10k.
⭐ Key Takeaways
The most critical concepts from this lecture are the ability to distinguish between scalars and vectors, and to perform basic vector algebraic operations. You must know the formulas for both the dot and cross products, including their component forms and geometric interpretations (dot product gives cosine of the angle, cross product gives sine and direction). Remember that a zero dot product implies perpendicular vectors, while a zero cross product implies parallel vectors. Finally, be prepared to calculate vector projections and the area of a parallelogram using the cross product magnitude.
🧠 Quick Revision Questions
- What is the formula for the unit vector in the direction of v?
- Given two vectors in component form, how do you find their sum and difference?
- Write the formula for the scalar (dot) product in terms of magnitudes and the angle between vectors.
- What is the condition for two non-zero vectors to be perpendicular?
- Using determinants, how do you compute the cross product of a = a₁i + a₂j + a₃k and b = b₁i + b₂j + b₃k?
📘 Lecture 11 — The Triple Scalar or Box Product
📖 Overview: This lecture introduces the triple scalar product (box product) of three vectors and its geometric interpretation as the volume of a parallelepiped. It then transitions to the concept of the gradient of a scalar function and the directional derivative, explaining how the gradient gives the direction of most rapid change. These concepts are fundamental in vector calculus and have crucial applications in physics, engineering, and optimization.
🗂️ Topics Covered
The lecture first covers the triple scalar product, also known as the box product, and derives its formula as the determinant of a 3x3 matrix formed by the components of the three vectors. It then introduces the gradient of a scalar function, the del operator, and the directional derivative, including its formula in terms of partial derivatives and its compact form using the gradient. The lecture concludes with the properties of directional derivatives, including the directions of most rapid increase, decrease, and zero change.
📝 Lecture Summary
The triple scalar or Box product
The product (a × b) . c is called the triple scalar product of a, b, and c (in that order). As | (a×b).c | = |a×b| |c| |cos θ|, the absolute value of the product is the volume of the parallelepiped (parallelogram-sided box) determined by a, b, and c. By treating the planes of b and c and of c and a as the base planes of the parallelepiped determined by a, b and c, we see that (a × b).c = (b×c).a = (c×a).b. Since the dot product is commutative, (a × b).c = a.(b×c).
For vectors a = a₁i + a₂j + a₃k, b = b₁i + b₂j + b₃k, and c = c₁i + c₂j + c₃k, the triple scalar product can be expressed as the determinant of a matrix: a . (b × c) = | a₁ a₂ a₃; b₁ b₂ b₃; c₁ c₂ c₃ |
🔑 Definition — Triple Scalar Product (Box Product): (a×b).c is the scalar product of the cross product of a and b with c.
📐 Formula: a . (b × c) = determinant | a₁ a₂ a₃; b₁ b₂ b₃; c₁ c₂ c₃ |
📌 Example: Find a.(b×c) for a = i + 2j – k, b = – 2i + 3k, c = 7j – 4k. a.(b×c) = | 1 2 -1; -2 0 3; 0 7 -4 | = 1 |0 3; 7 -4| – 2 |-2 3; 0 -4| + (-1) |-2 0; 0 7| = 1(0·(-4) – 3·7) – 2((-2)·(-4) – 3·0) + (-1)((-2)·7 – 0·0) = 1(0 – 21) – 2(8 – 0) + (-1)(-14 – 0) = -21 – 16 + 14 = -23 The volume is |a.(b×c)| = |-23| = 23. If we get a negative value, we take the Absolute value to make it positive, as volume is always positive.
💡 Why this matters: The triple scalar product provides a powerful computational tool to find the volume of a parallelepiped and can be used to test if three vectors are coplanar (volume = 0).
Gradient of a Scalar Function
∇ ≡ i ∂/∂x + j ∂/∂y + k ∂/∂z, ∇ is called the "del operator". The gradient of a scalar function φ is a vector operator defined as grad φ = (i ∂φ/∂x + j ∂/∂y + k ∂/∂z) φ = ∇φ. ∇φ is a vector quantity, and φ is a scalar quantity. Every component of ∇φ will operate with the function.
🔑 Definition — Gradient: The gradient of a scalar function φ is the vector operator ∇φ, which points in the direction of the greatest rate of increase of φ.
📝 Note: Gradient is a vector quantity.
Directional Derivative
If f(x,y) is differentiable at (x₀, y₀), and if u = (u₁, u₂) is a unit vector, then the directional derivative of f at (x₀, y₀) in the direction of u is defined by D_u f(x₀, y₀) = f_x(x₀, y₀)u₁ + f_y(x₀, y₀)u₂. It should be kept in mind that there are infinitely many directional derivatives of z = f(x,y) at a point (x₀, y₀), one for each possible choice of the direction vector u.
💡 Remarks (Geometrical interpretation): The directional derivative D_u f(x₀, y₀) can be interpreted algebraically as the instantaneous rate of change in the direction of u at (x₀, y₀) of z=f(x,y) with respect to the distance parameter s, or geometrically as the rise over the run of the tangent line to the curve C at the point Q₀.
🔑 Definition — Directional Derivative: The directional derivative measures the rate of change of a function in a specified direction.
📌 Example: Find the directional derivative of f(x, y) = 3x²y at the point (1, 2) in the direction of the vector a = 3i + 4j. f_x(x,y) = 6xy, f_y(x,y) = 3x², so f_x(1,2) = 12, f_y(1,2) = 3. The unit vector û = a/||a|| = (1/5)(3i + 4j) = (3/5)i + (4/5)j. D_u f(1,2) = 12(3/5) + 3(4/5) = 36/5 + 12/5 = 48/5.
📌 Example: Find the directional derivative of f(x, y) = 2x² + y² at the point P₀(-1, 1) in the direction of u = 3i – 4j. |u| = √(3² + (-4)²) = 5. Unit vector û = (3/5)i – (4/5)j. f_x = 4x, f_x(-1, 1) = -4. f_y = 2y, f_y(-1, 1) = 2. D_u f(-1,1) = f_x(-1,1)u₁ + f_y(-1,1)u₂ = (-4)(3/5) + (2)(-4/5) = -12/5 – 8/5 = -20/5 = -4.
💡 Remarks: If u = u₁i + u₂j is a unit vector making an angle θ with the positive x-axis, then u₁ = cos θ and u₂ = sin θ. The formula D_u f(x₀, y₀) = f_x(x₀, y₀)u₁ + f_y(x₀, y₀)u₂ can be written as D_u f(x₀, y₀) = f_x(x₀, y₀) cos θ + f_y(x₀, y₀) sin θ.
📝 Note: Formula for the directional derivative can be written in the following compact form using the gradient notation: D_u f(x, y) = ∇f(x, y) · û. The dot product of the gradient of f with a unit vector û produces the directional derivative of f in that direction.
📌 Example: Find the directional derivative of f(x,y) = e^(xy) at (-2, 0) in the direction of the unit vector u that makes an angle of π/3 with the positive x-axis. f_x(x,y) = y e^(xy), f_y(x,y) = x e^(xy). f_x(-2, 0) = 0, f_y(-2, 0) = -2. D_u f(-2, 0) = f_x(-2, 0) cos(π/3) + f_y(-2, 0) sin(π/3) = 0(1/2) + (-2)(√3/2) = -√3.
Gradient of function
If f is a function of x and y, then the gradient of f is defined as ∇f (x,y) = f_x(x,y)i + f_y(x,y)j. The directional derivative is D_u f(x, y) = ∇f (x, y) · û.
📌 Example: Find the directional derivative of f(x, y) = 2xy – 3y² at the point P₀(5, 5) in the direction of u = 4i + 3j. |u| = √(4² + 3²) = 5. Unit vector û = (4/5)i + (3/5)j. f_x = 2y, f_y = 2x – 6y. f_x(5,5) = 10, f_y(5,5) = 10 – 30 = -20. ∇f = 10i – 20j. D_u f(5,5) = ∇f · û = (10i – 20j) · ((4/5)i + (3/5)j) = 10(4/5) + (-20)(3/5) = 8 – 12 = -4.
📌 Example: Find the directional derivative of f(x,y) = xe^y + cos(xy) at the point (2, 0) in the direction of a = 3i – 4j. û = (3/5)i – (4/5)j. f_x(x,y) = e^y – y sin(xy), f_y(x,y) = xe^y – x sin(xy). f_x(2,0) = e⁰ – 0 = 1, f_y(2,0) = 2e⁰ – 2(0) = 2. ∇f(2,0) = i + 2j. D_u f(2,0) = (i + 2j) · ((3/5)i – (4/5)j) = 1(3/5) + 2(-4/5) = 3/5 – 8/5 = -5/5 = -1.
Properties of Directional Derivatives
D_u f = ∇f · û = |∇f| cos θ
- The function f increases most rapidly when cos θ = 1, or when û is the direction of ∇f. That is, at each point P in its domain, f increases most rapidly in the direction of the gradient vector ∇f at P. The derivative in this direction is D_u f = |∇f| cos(0) = |∇f|.
- Similarly, f decreases most rapidly in the direction of –∇f. The derivative in this direction is D_u f = |∇f| cos(π) = – |∇f|.
- Any direction û orthogonal to the gradient is a direction of zero change in f, because cos θ then equals π/2 and D_u f = |∇f| cos(π/2) = |∇f|·0 = 0.
📌 Example: For f(x, y) = x²/2 + y²/2 at (1,1): a) The direction of rapid change: The function increases most rapidly in the direction of ∇f at (1,1). ∇f(1,1) = (xi + yj)(1,1) = i + j. Its direction is û = (i + j)/|i + j| = (i + j)/√(1² + 1²) = (1/√2)i + (1/√2)j. b) The directions of zero change: The directions of zero change at (1,1) are the directions orthogonal to ∇f. n = (-i + j)/√2 and –n = (i – j)/√2.
⭐ Key Takeaways
The triple scalar product a·(b×c) gives the signed volume of a parallelepiped, and its absolute value is the volume. It can be computed as the determinant of the matrix formed by the components of the three vectors. The gradient ∇f is a vector operator that points in the direction of the most rapid increase of a scalar function f at a given point, and its magnitude |∇f| is the rate of that increase. The directional derivative D_u f measures the rate of change of f in the direction of a unit vector û and can be computed as the dot product of the gradient and the unit direction vector. The directional derivative is maximized when û is parallel to ∇f (direction of steepest ascent), minimized when opposite to ∇f (steepest descent), and zero when perpendicular to ∇f.
🧠 Quick Revision Questions
- How do you compute the volume of the parallelepiped formed by three vectors a, b, and c?
- What is the relationship between the triple scalar products (a×b)·c, (b×c)·a, and (c×a)·b?
- Write the expression for the directional derivative of f(x,y) at (x₀, y₀) in the direction of a unit vector u = (u₁, u₂).
- If ∇f at a point is the zero vector, what does this imply about the directional derivatives in all directions?
- In which direction does a function f increase most rapidly at a given point, and what is the value of the maximum directional derivative?
📘 Lecture 12 — Tangent planes to the surfaces
📖 Overview: This lecture introduces the concept of tangent planes and normal lines to surfaces in three-dimensional space. It builds on the definition of tangent lines to curves and extends it to surfaces, covering the necessary prerequisite knowledge of lines and planes before deriving the equations for tangent planes and normal lines using gradients.
🗂️ Topics Covered
The lecture begins with definitions of tangent planes and surface normals, then reviews different forms of equations of straight lines in two and three dimensions, parametric equations, and implicit vs. explicit forms of curves. It covers the equation of a plane in point-normal form, the relationship between gradients and tangents to surfaces, and finally derives the equations for the tangent plane and normal line to a surface defined by f(x,y,z)=0, with worked examples.
📝 Lecture Summary
Tangent planes to the surfaces / Normal line to the surfaces
If C is a smooth parametric curve in three dimensions, then the tangent line to C at point P0 is the line through P0 along the unit tangent vector to C at P0. The concept of a tangent plane builds on this definition. If P0(x0,y0,z0) is a point on the surface S, and if the tangent lines at P0 to all smooth curves that pass through P0 and lie on S all lie in a common plane, then that plane is regarded as the tangent plane to the surface S at P0. Its normal (the straight line through P0 perpendicular to the tangent) is called the surface normal of S at P0.
Different forms of equation of straight line in two dimensional space
- Slope intercept form: y = mx + c, where m is the slope and c is the y-intercept.
- Point-Slope Form: If m is the slope and P0(x0,y0) is a point on the line, then y – y0 = m (x – x0). Alternatively, the rise/run relationship gives: (y – y0)/b = (x – x0)/a, leading to y – y0 = (b/a)(x – x0).
- General Equation: Ax + By + C = 0, where m = -A/B.
- Parametric equation of a line: In two dimensions, the parametric equations passing through (x0,y0) and parallel to vector ai + bj are: x = x0 + at, y = y0 + bt. Eliminating t gives (x – x0)/a = (y – y0)/b, or y – y0 = (b/a)(x – x0). The parametric vector form is r(t) = (x0 + at)i + (y0 + bt)j.
Equation of line in three dimensional
Parametric equations of a line in three-dimensional space passing through (x0,y0,z0) and parallel to vector ai + bj + ck are: x = x0 + at, y = y0 + bt, z = z0 + ct. Eliminating t gives (x – x0)/a = (y – y0)/b = (z – z0)/c.
📌 Example: Find parametric equations for the straight line through point A(2,4,3) parallel to vector v = 4i + 0j – 7k. Here x0=2, y0=4, z0=3 and a=4, b=0, c=-7. The required parametric equations are: x = 2 + 4t, y = 4 + 0t, z = 3 – 7t.
Different forms of equations of curves
Curves in the plane are defined in different ways:
- Explicit form: y = f(x). Example: y = √(9 – x²), -3 ≤ x ≤ 3.
- Implicit form: F(x,y) = 0. Example: x² + y² = 9, -3 ≤ x ≤ 3, 0 ≤ y ≤ 3.
- Parametric form: x = f(t), y = g(t). Example: x = 3cosθ, y = 3sinθ, 0 ≤ θ ≤ π. This satisfies x² + y² = 9cos²θ + 9sin²θ = 9.
- Parametric vector form: r(t) = f(t)i + g(t)j, a ≤ t ≤ b. Example: r(θ) = 3cosθ i + 3sinθ j, 0 ≤ θ ≤ π.
Equation of a plane
A plane can be completely determined if we know one point on it and the direction of the perpendicular (normal) to it. Let a plane pass through point P0(x0,y0,z0) with normal vector n = ai + bj + ck. Let P(x,y,z) be any point on the plane. Then vector P0P = (x – x0)i + (y – y0)j + (z – z0)k lies in the plane, and n is perpendicular to P0P. Since n ⊥ P0P, the dot product is zero: n · P0P = 0. This gives the point-normal form of the equation of a plane: a(x – x0) + b(y – y0) + c(z – z0) = 0. This can be rewritten as ax + by + cz + d = 0, where d = –ax0 – by0 – cz0.
📌 Example: Find an equation of the plane passing through (3,-1,7) perpendicular to vector n = 4i + 2j – 5k. Using point-normal form: 4(x – 3) + 2(y + 1) – 5(z – 7) = 0. Simplifying: 4x – 12 + 2y + 2 – 5z + 35 = 0 → 4x + 2y – 5z + 25 = 0.
🔑 Definition — Relationship between gradient and plane equation: For the general equation of a straight line ax + by + c = 0, if φ(x,y) = ax + by, then φx = a, φy = b, and ∇φ = ai + bj = n. The gradient vector ∇φ is always normal to the plane. Similarly, for a plane ax + by + cz + d = 0, if φ = ax + by + cz, then φx = a, φy = b, φz = c, and ∇φ = ai + bj + ck, which is always normal to the plane.
Gradients and Tangents to Surfaces
For a differentiable function f(x,y) having a constant value c along a smooth curve, the curve has parametric equation x = g(t), y = h(t), r = g(t)i + h(t)j. Differentiating f(g(t), h(t)) = c with respect to t using the chain rule: (∂f/∂x)(dg/dt) + (∂f/∂y)(dh/dt) = 0. This gives: (∂f/∂x i + ∂f/∂y j) · (dg/dt i + dh/dt j) = 0, i.e., ∇f · (dr/dt) = 0. Therefore, ∇f is normal to the tangent vector dr/dt, so it is normal to the curve through (x0,y0).
💡 Why this matters: This relationship between the gradient and the tangent vector is the key to finding tangent planes.
Tangent Plane and Normal Line
Consider all curves through point P0(x0,y0,z0) on a surface f(x,y,z) = 0. The plane containing all tangents to these curves at P0 is called the tangent plane to the surface at P0. The straight line perpendicular to all these tangent lines at P0 is called the normal line to the surface at P0. If fx, fy, fz are all continuous at P0 and not all zero, then the gradient ∇f = fxi + fyj + fzk at P0 gives the direction of the normal vector to the surface at P0.
🔑 Definition — Equation of Tangent Plane: If P0(x0,y0,z0) is any point on the surface f(x,y,z) = 0, and if f(x,y,z) is differentiable at P0, then the tangent plane at P0 has the equation: fx(x0,y0,z0)(x – x0) + fy(x0,y0,z0)(y – y0) + fz(x0,y0,z0)(z – z0) = 0.
📌 Example 1: Find the tangent plane to 9x² + 4y² – z² = 36 at P(2,3,6). f(x,y,z) = 9x² + 4y² – z² – 36 fx = 18x, fy = 8y, fz = –2z At P: fx(P) = 36, fy(P) = 24, fz(P) = –12 Equation: 36(x – 2) + 24(y – 3) – 12(z – 6) = 0 Dividing by 12: 3x – 6 + 2y – 6 – z + 6 = 0 → 3x + 2y – z – 6 = 0
📌 Example 2: Find the tangent plane to z = x cos y – yex at (0,0,0). Rewrite as: x cos y – yex – z = 0, so f(x,y,z) = x cos y – yex – z fx(x,y) = cos y – yex, so fx(0,0,0) = 1 – 0(1) = 1 fy(x,y) = –x sin y – ex, so fy(0,0,0) = 0 – 1 = –1 fz = –1, so fz(0,0,0) = –1 Equation: 1(x – 0) + (–1)(y – 0) + (–1)(z – 0) = 0 → x – y – z = 0
⭐ Key Takeaways
The gradient vector ∇f at a point on a surface f(x,y,z)=0 is always normal to the surface at that point, making it the key to finding both tangent planes and normal lines. The tangent plane equation is derived from the dot product of the gradient (normal vector) with any vector lying in the plane. Parametric and point-normal forms of lines and planes are essential prerequisite concepts. For exam problems, always rewrite the surface equation as f(x,y,z)=0 before computing partial derivatives, and evaluate all partial derivatives at the given point before writing the final plane equation.
🧠 Quick Revision Questions
- What is the condition for a plane to be considered the tangent plane to a surface at a point P0?
- Write the point-normal form of the equation of a plane.
- For a surface f(x,y,z)=0, what does the gradient vector ∇f at a point represent?
- Find the equation of the tangent plane to z = x² + y² at the point (1,1,2).
- A line passes through (1,2,3) parallel to the vector 2i - j + 4k. Write its parametric equations.